2024-03-29T10:34:12Z
https://zenodo.org/oai2d
oai:zenodo.org:5801419
2021-12-27T10:13:29Z
user-sustage
Georgios Athanassiou
Maria Pateraki
Iraklis Varlamis
2021-12-23
<p>Abstract</p>
<p>The paper outlines the sustAGE system, a smart solution that builds upon strategic technology trends, such as Internet-of-Things, machine learning and recommender systems, to support sustainable work environments and increase wellness at work and well-being with a focus on the ageing workforce. Acknowledging the interrelation of the work and private arrays for healthy ageing, the developed solution utilizes a recommendation-based approach providing personalized warnings and preventive recommendations regarding occupational risks, as well as personalized cognitive and physical training activities for the off-work context with the overall goal of maintaining Work Ability and enabling sustainable work. The piloting of the proposed solution in two critical industrial domains provides promising results towards the use of personalized recommendation-based interventions for the working context and beyond for improving workers’ occupational safety and health, performance and general well-bei ng.</p>
https://doi.org/10.5220/0010723000003063
oai:zenodo.org:5801419
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/restrictedAccess
IJCII 2021, 13th International Joint Conference on Computational Intelligence, 27 OCTOBER 2021
Recommendation Systems
Micro-moments
Occupational Safety and Health
Well-being
Work Ability
Sustainable Work
Recommendation-based Interventions
Micro-moment-based Interventions for a Personalized Support of Healthy and Sustainable Ageing at Work: Development and Application of a Context-sensitive Recommendation Framework
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:5801443
2022-06-03T11:43:45Z
user-sustage
Mónica Muiños
Soledad Ballesteros
2021-02-01
<p>Background</p>
<p>Dance is a multidomain activity that combines aerobic, coordination and cognitive exercise. This music-associated physical and cognitive exercise is a leisure activity that motivates people, elicits emotions, and avoids boredom, promoting adherence to practice. Continuing physical activity is of paramount importance, since cognitive benefits tend to disappear or even reverse when training ceases.</p>
<p>Objective</p>
<p>The question we addressed in this systematic review is what influence dance has on the brain and cognition of healthy middle-aged and older adults.</p>
<p>Literature survey</p>
<p>We systematically reviewed the effects of dance on brain and cognition in older adults using MEDLINE, Psyc-Info, PubMed and Scopus databases.</p>
<p>Methodology</p>
<p>After screening 1051 studies, thirty-five met the eligibility inclusion criteria. These studies showed that dance improves brain structure and function as well as physical and cognitive functions.</p>
<p>Conclusions</p>
<p>The protective effect of dance training on cognition in older adults, together with the possibility of adapting intensity and style to suit possible physical limitations makes this activity very suitable for older adults.</p>
https://doi.org/10.1016/j.neubiorev.2020.11.028
oai:zenodo.org:5801443
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/restrictedAccess
PUBMED Neuroscience & Biobehavioral Reviews, 121(February 2021), 259-276, (2021-02-01)
Aging
Dance Music
Neuroplasticity
Physical exercise
Does dance counteract age-related cognitive and brain declines in middle-aged and older adults? A systematic review
info:eu-repo/semantics/article
oai:zenodo.org:5801380
2021-12-27T13:48:40Z
user-sustage
Adria Mallol-Ragolta
Helena Cuesta
Emilia Gómez
Björn W. Schuller
2021-08-30
<p>Abstract</p>
<p>The aim of this contribution is to automatically detect COVID-19 patients by analysing the acoustic information embedded in coughs. COVID-19 affects the respiratory system, and, consequently, respiratory-related signals have the potential to contain salient information for the task at hand. We focus on analysing the spectrogram representations of cough samples with the aim to investigate whether COVID-19 alters the frequency content of these signals. Furthermore, this work also assesses the impact of gender in the automatic detection of COVID-19. To extract deep-learnt representations of the spectrograms, we compare the performance of a cough-specific, and a Resnet18 pre-trained Convolutional Neural Network (CNN). Additionally, our approach explores the use of contextual attention, so the model can learn to highlight the most relevant deep-learnt features extracted by the CNN. We conduct our experiments on the dataset released for the Cough Sound Track of the DICOVA 2021 Challenge. The best performance on the test set is obtained using the Resnet18 pre-trained CNN with contextual attention, which scored an Area Under the Curve (AUC) of 70.91% at 80% sensitivity.</p>
https://doi.org/10.21437/Interspeech.2021-1052
oai:zenodo.org:5801380
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
INTERSPEECH 2021, International Speech Communication Association 2021, Brno Czechia, 30 August 2021- 3 September 2021
COVID-19
Neural Network
Gender Information
Cough-Based COVID-19 Detection with Contextual Attention Convolutional Neural Networks and Gender Information
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6610459
2022-06-03T13:50:42Z
user-sustage
Emilia Parada-Cabaleiro
Anton Batliner
Markus Schedl
2022-01-16
<p>Abstract</p>
<p>Musical listening is broadly used as an inexpensive and safe method to reduce self-perceived anxiety. This strategy is based on the emotivist assumption claiming that emotions are not only recognised in music but induced by it. Yet, the acoustic properties of musical work capable of reducing anxiety are still under-researched. To fill this gap, we explore whether the acoustic parameters relevant in music emotion recognition are also suitable to identify music with relaxing properties. As an anxiety indicator, the positive statements from the six-item Spielberger State-Trait Anxiety Inventory, a self-reported score from 3 to 12, are taken. A user-study with 50 participants assessing the relaxing potential of four musical pieces was conducted; subsequently, the acoustic parameters were evaluated. Our study shows that when using classical Western music to reduce self-perceived anxiety, tonal music should be considered. In addition, it also indicates that harmonicity is a suitable indicator of relaxing music, while the role of scoring and dynamics in reducing non-pathological listener distress should be further investigated.</p>
https://doi.org/10.5281/zenodo.6610459
oai:zenodo.org:6610459
eng
Zenodo
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.6610458
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
International Journal of Environmental Research and Public Health, 19(2), (2022-01-16)
music psychology
audio features
self-report
signal processing
induced distress
every-day anxiety
An Exploratory Study on the Acoustic Musical Properties to Decrease Self-Perceived Anxiety
info:eu-repo/semantics/article
oai:zenodo.org:6610084
2022-06-03T13:50:44Z
user-sustage
user-eu
Manolis Lourakis
George Terzakis
2021-05-05
<p><strong>Abstract:</strong></p>
<p>The perspective-n-point (PnP) problem is of fundamental importance in computer vision. A global optimality condition for PnP that is independent of a particular rotation parameterization was recently developed by Nakano. This paper puts forward a direct least squares, algebraic PnP solution that extends Nakano's work by combining his optimality condition with the modified Rodrigues parameters (MRPs) for parameterizing rotation. The result is a system of polynomials that is solved using the Gröbner basis approach. An MRP vector has twice the rotational range of the classical Rodrigues (i.e., Cayley) vector used by Nakano to represent rotation. The proposed solution provides strong guarantees that the full rotation singularity associated with MRPs is avoided. Furthermore, detailed experiments provide evidence that our solution attains accuracy that is indistinguishable from Nakano's Cayley-based method with a moderate increase in computational cost.</p>
https://doi.org/10.5281/zenodo.6610084
oai:zenodo.org:6610084
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610083
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
ICPR, 2020 25th International Conference on Pattern Recognition (ICPR), Mialn, Italy, 10-15 January 2021
A Globally Optimal Method for the PnP Problem with MRP Rotation Parameterization
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4294150
2020-11-28T12:27:15Z
user-sustage
user-eu
Ziping Zhao
Zhongtian Bao
Zixing Zhang
Nicholas Cummins
Haishuai Wang
Björn W. Schuller
2019-09-19
<p>Discrete speech emotion recognition (SER), the assignment of a single emotion label to an entire speech utterance, is typically performed as a sequence-to-label task. This approach, however, is limited, in that it can result in models that do not capture temporal changes in the speech signal, including those indicative of a particular emotion. One potential solution to overcome this limitation is to model SER as a sequence-to-sequence task instead. In this regard, we have developed an attention-based bidirectional long short-term memory (BLSTM) neural network in combination with a connectionist temporal classification (CTC) objective function (Attention-BLSTM-CTC) for SER. We also assessed the benefits of incorporating two contemporary attention mechanisms, namely component attention and quantum attention, into the CTC framework. To the best of the authors’ knowledge, this is the first time that such a hybrid architecture has been employed for SER.We demonstrated the effectiveness of our approach on the Interactive Emotional Dyadic Motion Capture (IEMOCAP) and FAU-Aibo Emotion corpora. The experimental results demonstrate that our proposed model outperforms current state-of-the-art approaches.</p>
The work presented in this paper substantially supported by the National Natural Science Foundation of China (Grant No. 61702370), the Key Program of the Natural Science Foundation of Tianjin (Grant No. 18JCZDJC36300), the Open Projects Program of the National Laboratory of Pattern Recognition, and the Senior Visiting Scholar Program of Tianjin Normal University.
Interspeech 2019
ISSN: 1990-9772
https://doi.org/10.21437/Interspeech.2019-1649
oai:zenodo.org:4294150
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
INTERSPEECH 2019, 20th Annual Conference of the International Speech Communication Association, Graz, Austria, 15-19 September 2019
speech emotion recognition, connectionist tem- poral classification, attention mechanism, bidirectional LSTM
Attention-enhanced Connectionist Temporal Classification for Discrete Speech Emotion Recognition
info:eu-repo/semantics/article
oai:zenodo.org:4294283
2020-12-08T14:35:14Z
user-sustage
Eloisa Ruiz-Marquez
Antonio Prieto
Julia Mayas
Pilar Toril
José Manuel Reales
Soledad Ballesteros
2019-11-26
<p>Abstract</p>
<p><strong><em>Objective:</em></strong> In this intervention study, we investigated the benefits of nonaction videogames on measures of selective attention and visuospatial working memory (WM) in young adults.</p>
<p><strong><em>Materials and Methods:</em></strong> Forty-eight young adults were randomly assigned to the experimental group or to the active control group. The experimental group played 10 nonaction adaptive videogames selected from <em>Lumosity</em>, whereas the active control group played two nonadaptive simulation-strategy games (<em>SimCity</em> and <em>The Sims</em>). Participants in both groups completed 15 training sessions of 30 minutes each. The training was conducted in small groups. All the participants were tested individually before and after training to assess possible transfer effects to selective attention, using a Cross-modal Oddball task, inhibition with the Stroop task, and visuospatial WM enhancements with the Corsi blocks task.</p>
<p><strong><em>Results:</em></strong> Participants improved videogame performance across the training sessions. The results of the transfer tasks show that the two groups benefited similarly from game training. They were less distracted and improved visuospatial WM.</p>
<p><strong><em>Conclusion:</em></strong> Overall, there was no significant interaction between group (group trained with adaptive nonaction videogames and the active control group that played simulation games) and session (pre- and post-assessment). As we did not have a passive nonintervention control group, we cannot conclude that adaptive nonaction videogames had a positive effect, because some external factors might account for the pre- and post-test improvements observed in both groups.</p>
https://doi.org/10.1089/g4h.2019.0004
oai:zenodo.org:4294283
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/restrictedAccess
Games for Health Journal, VOL. 8, NO. 6, (2019-11-26)
Effects of Nonaction Videogames on Attention and Memory in Young Adults
info:eu-repo/semantics/article
oai:zenodo.org:7092270
2022-09-20T02:26:33Z
user-sustage
Iraklis Varlamis
Michail Maniadakis
Georgios Athanassiou
2022-09-19
<p>Today, a large part of the labor policies in the EU aim at extending the active partici- pation of older (i.e. 50+) employees in the workforce in order to avoid the respective pressure on the national economies and health systems as well as potential shortco- mings in qualified personnel due to demographical changes in the entire population. Preventing involuntary early retirement goes hand in hand with supporting self- sufficient and healthy living. The present work considers the use and exploitation of modern technological advancements to support the achievement of the above goal. Specifically, we propose a new approach to developing complex recommendation systems, which are capable of monitoring and supporting the daily activities of emplo- yees in a personalized manner, both at work and during their broader daily activities. The proposed approach is based on the new Micro-Moments (MiMos) concept for cri- tical event recognition, incorporating multiple streams of complementary information from a distributed sensor network that is flowing into the system based on IoT techno- logies. The recommendation system follows a user-centered approach for providing (personalized) suggestions that support the occupational safety of users, improve their health and enhance their productivity, in a personalized way. This paper summarizes the concept of Micro-Moments (MiMos) and how it contributes to issuing recom- mendations based on specific user needs. We also present the current version and implementation of the system in the field of port logistics, where it is observed that recommendations delivered at the right time to the right person can help improve the efficiency of the workforce and extend its working capacity.</p>
https://doi.org/10.54941/ahfe1002144
oai:zenodo.org:7092270
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
Human Factors and Systems Interaction, (2022-09-19)
AHFE 2022, International Conference on Applied Human Factors and Ergonomics, New York, USA, 24-28 July 2022
Recommender systems
Human systems integration
Occupational safety and health
A Micro-Moment Recommendation Framework in Industrial Environments
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:5801429
2021-12-27T09:38:18Z
user-sustage
user-eu
Konstantinos Papoutsakis
Thodoris Papadopoulos
Michalis Maniadakis
Manolis Lourakis
Maria Pateraki
Iraklis Varlamis
2021-07-13
<p>Abstract</p>
<p>sustAGE is an ongoing project, developing an Internet of Things ecosystem, including smartphones, smartwatches, localization and environmental sensors and cameras to support ageing workers in industrial environments while performing assembly tasks, such a car manufacturing factory. In this context, we briefly describe a non-obtrusive method for assessing the physical strain of workers using visual data and a method for detecting worker fatigue using heart rate data acquired by smartwatches. The results of both methods are utilized by the recommendation system developed in sustAGE to support preventive actions towards work-related musculo-skeletal disorders and fatigue and to promote occupational safety.</p>
<p> </p>
https://doi.org/10.1145/3453892.3461633
oai:zenodo.org:5801429
eng
Association for Computing MachineryNew YorkNYUnited States
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/restrictedAccess
PETRA 2021, The 14th PErvasive Technologies Related to Assistive Environments Conference, Corfu, Greece, 29 June 2021 -2July 2021
Detection of physical strain and fatigue in industrial environments using visual and non-visual sensors
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6610418
2022-06-03T10:27:53Z
user-sustage
user-eu
Adria Mallol-Ragolta
Anastasia Semertzidou
Maria Pateraki
Björn Schuller
2022-06-03
<p><strong>Abstract:</strong></p>
<p>This work introduces the harAGEdataset: a novel multimodal smartwatch-based dataset for Human Activity Recognition (HAR) with more than 17 hours of data collected from 19 participants using a Garmin Vivoactive 3 device. The dataset contains samples from resting, lying, sitting, standing, washing hands, walking, running, stairs climbing, strength workout, flexibility workout, and cycling activities. The resting activity, excluded from the set of activities to recognise, was explicitly conducted while avoiding stressors and external stimuli, so the data collected can be used to compute the personal, baseline heart rate at rest. We also present the HAR-based models trained using the accelerometer data to recognise different sets of activities. Specifically, we focus on different strategies to combine, fuse, and enrich the accelerometer measurements, so they can be used end-to-end. Model performances are assessed following a Leave-One-Subject-Out Cross-Validation (LOSO-CV) approach, and we use the Unweighted Average Recall (UAR) as the evaluation metric to compare the ground truth and the inferred information. The best UAR score of 98.1 % is obtained when recognising the static and the dynamic activities, excluding the samples corresponding to the washing hands, strength workout, and flexibility workout activities. When recognising the specific activities included in these two sets, the model with the best performance scores a UAR of 70.1 %. Finally, when recognising all the activities considered in the harAGEdataset, the highest UAR achieved is 64.3 %.</p>
https://doi.org/10.5281/zenodo.6610418
oai:zenodo.org:6610418
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610417
info:eu-repo/semantics/restrictedAccess
FG 2021, 2021 16th IEEE International Conference on Automatic Face and Gesture Recognition, Jodhpur, India, 15-18 December 2021
Accelerometers
Fuses
Computational modeling
Neural networks
Network architecture
Activity recognition
Stairs
harAGE: A Novel Multimodal Smartwatch-based Dataset for Human Activity Recognition
info:eu-repo/semantics/article
oai:zenodo.org:4106475
2020-12-09T11:31:58Z
user-sustage
user-eu
Maximilian Schmitt
Nicholas Cummins
Björn W. Schuller
2019-09-19
<p>Emotion recognition in speech is a meaningful task in affective computing and human-computer interaction. As human emotion is a frequently changing state, it is usually represented as a densely sampled time series of emotional dimensions, typically arousal and valence. For this, recurrent neural network (RNN) architectures are employed by default when it comes to modelling the contours with deep learning approaches. However, the amount of temporal context required is questionable, and it has not yet been clarified whether the consideration of long-term dependencies is actually beneficial. In this contribution, we demonstrate that RNNs are not necessary to accomplish the task of time-continuous emotion recognition. Indeed, results gained indicate that deep neural networks incorporating less complex convolutional layers can provide more accurate models. We highlight the pros and cons of recurrent and non-recurrent approaches and evaluate our methods on the public SEWA database, which was used as a benchmark in the 2017 and 2018 editions of the Audio-Visual Emotion Challenge.</p>
ISSN: 1990-9772, Pages 2808-2812
https://doi.org/10.21437/Interspeech.2019-2710
oai:zenodo.org:4106475
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
INTERSPEECH 2019, 20th Annual Conference of the International Speech Communication Association, Graz, Austria, September 15–19, 2019
affective computing, speech emotion recogni- tion, human-computer interaction, computational paralinguis- tics, convolutional neural networks
Continuous Emotion Recognition in Speech – Do We Need Recurrence?
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4294162
2020-12-03T11:30:13Z
user-sustage
user-eu
Adria Mallol-Ragolta
Maximilian Schmitt
Alice Baird
Nicholas Cummins
Björn Schuller
2020-07-11
<p>The human ability to empathise is a core aspect of successful interpersonal relationships. In this regard, human-robot interaction can be improved through the automatic perception of empathy, among other human attributes, allowing robots to affectively adapt their actions to interactants' feelings in any given situation. This paper presents our contribution to the generalised track of the One-Minute Gradual (OMG) Empathy Prediction Challenge by describing our approach to predict a listener's valence during semi-scripted actor-listener interactions. We extract visual and acoustic features from the interactions and feed them into a bidirectional long short-term memory network to capture the time-dependencies of the valence-based empathy during the interactions. Generalised and personalised unimodal and multimodal valence-based empathy models are then trained to assess the impact of each modality on the system performance. Furthermore, we analyse if intra-subject dependencies on empathy perception affect the system performance. We assess the models by computing the concordance correlation coefficient (CCC) between the predicted and self-annotated valence scores. The results support the suitability of employing multimodal data to recognise participants' valence-based empathy during the interactions, and highlight the subject-dependency of empathy. In particular, we obtained our best result with a personalised multimodal model, which achieved a CCC of 0.11 on the test set.</p>
Funding : Bavarian State Ministry of Education, Science and the Arts
Electronic ISBN:978-1-7281-0089-0
IEEE restrictions of institutional members and purchases
https://doi.org/10.1109/FG.2019.8756517
oai:zenodo.org:4294162
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/restrictedAccess
FG2019, 14th IEEE International Conference on Automatic Face and Gesture Recognition, Performance Analysis of Unimodal and Multimodal Models in Valence-Based Empathy Recognition, Lille, France, 14-18 May 2019
emotion recognition
feature extraction
human-robot interaction
Performance Analysis of Unimodal and Multimodal Models in Valence-Based Empathy Recognition
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6610377
2022-06-03T13:50:41Z
user-sustage
user-eu
Adria Mallol-Ragolta
Shuo Liu
Björn W. Schuller
2021-12-09
<p><strong>Abstract:</strong></p>
<p>Face masks alter the speakers’ voice, as their intrinsic properties provide them with acoustic absorption capabilities. Hence, face masks act as filters to the human voice. This work focuses on the automatic detection of face masks from speech signals, emphasising on a previous work claiming that face masks attenuate frequencies above 1 kHz. We compare a paralinguistics-based and a spectrograms-based approach for the task at hand. While the former extracts paralinguistic features from filtered versions of the original speech samples, the latter exploits the spectrogram representations of the speech samples containing specific ranges of frequencies. The machine learning techniques investigated for the paralinguistics-based approach include Support Vector Machines (SVM), and a Multi-Layer Perceptron (MLP). For the spectrograms-based approach, we use a Convolutional Neural Network (CNN). Our experiments are conducted on the Mask Augsburg Speech Corpus (MASC), released for the Interspeech 2020 Computational Paralinguistics Challenge (COMPARE). The best performances on the test set from the paralinguistic analysis are obtained using the high-pass filtered versions of the original speech samples. Nonetheless, the highest Unweighted Average Recall (UAR) on the test set is obtained when exploiting the spectrograms with frequency content below 1 kHz.</p>
https://doi.org/10.5281/zenodo.6610377
oai:zenodo.org:6610377
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610376
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
EMBC, 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society, Mexico, 01-05 November 2021
Support vector machines
Training
Absorption
Transfer learning
Surgery
Feature extraction
Convolutional neural networks
The Filtering Effect of Face Masks in their Detection from Speech
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4304888
2020-12-08T11:33:28Z
user-sustage
Decky Aspandi
Adria Mallol-Ragolta
Bj ̈orn Schuller
Xavier Binefa
2020-12-01
<p>There is a growing interest in affective computing research nowadays given its crucial role in bridging humans with computers. This progress has recently been accelerated due to the emergence of bigger dataset. One recent advance in this field is the use of adversarial learning to improve model learning through augmented samples. However, the use of latent features, which is feasible through adversarial learning, is not largely explored, yet. This technique may also improve the performance of affective models, as analogously demonstrated in related fields, such as computer vision. To expand this analysis, in this work, we explore the use of latent features through our proposed adversarial-based networks for valence and arousal recognition in the wild. Specifically, our models operate by aggregating several modalities to our discriminator, which is further conditioned to the extracted latent features by the generator. Our experiments on the recently released SEWA dataset suggest the progressive improvements of our results. Finally, we show our competitive results on the Affective Behavior Analysis in-the-Wild (ABAW) challenge dataset.</p>
https://arxiv.org/abs/2002.00883
https://doi.org/10.1109/FG47880.2020.00053
oai:zenodo.org:4304888
eng
IEEE
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
FG 2020, 15th IEEE International Conference on Automatic Face and Gesture Recognition, Buenos Aires, AR, 16 - 20 May
Computer Vision and Pattern Recognition (cs.CV);
Machine Learning (cs.LG)
Latent-Based Adversarial Neural Networks for Facial Affect Estimations
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:7092317
2022-09-20T02:26:30Z
user-sustage
Konstantinos Papoutsakis
Manolis Lourakis
Maria Pateraki
2022-09-19
<p>Abstract: The aim of this study is to investigate the development and the evaluation of a computer vision-based framework to aid the automatic assessment of posture deviations in assembly tasks in realistic work environments. A posture deviation refers to a time-varying working posture performed by the worker, that deviates from ergonomically safe body postures expected in the context of particular work tasks and is known to impose increased physical strain. The estimation of their occurrences can serve as indicators, known as risk factors, for the assessment of physical ergonomics towards the prevention of physical strain and in the-long-term of work-related musculo-skeletal disorders (WMSD). Using visual information acquired by camera sensors, our goal is to estimate the full body motion of a line worker in 3D space, unobtrusively, and to perform classification of four types of posture deviations, also noted as ergonomically sub-optimal working postures that were employed by the MURI risk analysis tool. We formulate a learning-based action classification task using Deep Graph-based Neural Networks and differential temporal alignment cost as a classification measure to estimate the type and risk level of the observed posture deviation during work activities. To evaluate the efficiency of the proposed approach, a new video dataset was captured in the context of the sustAGE project, that demonstrate two different workers during car door assembly actions in a simulated production line in an actual workplace. Rich annotation data were provided by experts in manufacturing and ergonomics. Both quantitative and qualitative evaluation of the proposed framework provide evidence for its efficiency and reliability in supporting ergonomic risk assessment and preventive actions for WMSD in real working environments.</p>
https://doi.org/10.54941/ahfe1002145
oai:zenodo.org:7092317
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
AHFE 2022, International Conference on Applied Human Factors and Ergonomics, New York, USA, 24-28 July 2022
occupational safety
physical strain
ergonomic risk assessment
posture devations
computer vision
3D human pose estimation
manufacturing
Automatic assessment of posture deviations in assembly tasks
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:5801406
2022-01-07T12:47:13Z
user-sustage
Soledad Ballesteros
Grzegorz Sedek
Thomas Hess
Dayna Touron
2021-11-22
<p>The study of aging and cognition has grown exponentially over the past 50 years, developing from a field dominated by experimentally based information-processing traditions to one represented by a more mature approach both conceptually and methodologically. In the past 10 years there has been growth in integrative approaches that incorporate behavioral, neuropsychological, and social information. In addition, there has been a growing recognition of the limitations associated with simple cross-sectional age-group comparisons, along with an increased use of more complex methods. This has resulted in the development of increasingly sophisticated research designs and analytic tools focused on understanding a multitude of potential mediators and moderators of cognitive change. The result has been a move away from negative-views of cognitive aging to one that is more nuanced and sensitive to contextual factors.<br>
<br>
<em>Multiple Pathways of Cognitive Aging</em> explores the factors associated with adaptive functioning in later life. Its emphasis is on understanding both the factors underlying individual differences in change in cognitive functioning in later life and the nature of the compensatory mechanisms developed by most successful and active middle-aged and older adults. This includes a consideration of motivational factors as a driver of both cognitive change and adaptive functioning.<br>
<br>
For students and researchers, <em>Multiple Pathways of Cognitive Aging </em>offers valuable insights into the field of cognitive development, along with innovative methodological approaches to help them in their own research.</p>
https://doi.org/10.5281/zenodo.5801406
oai:zenodo.org:5801406
Oxford University Press
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.5801405
info:eu-repo/semantics/restrictedAccess
cognitive aging
implicit memory
adaptive functioning
cognitive health
cognitive training
dance
decision making
engagement
episodic memory
executive functions
multidomain training
neuroplasticity
video games
Cognitive plasticity induced in older adults by cognitive training, physical exercise, and combined interventions, Chapter 14
info:eu-repo/semantics/bookPart
oai:zenodo.org:4294181
2020-12-08T11:44:41Z
user-sustage
user-eu
Shahin Amiriparian
Jing Han
Maximilian Schmitt
Alice Baird
Adria Mallol-Ragolta
Manuel Milling
Maurice Gerczuk
Björn Schulle
2020-11-27
<p>Abstract:</p>
<p>During both positive and negative dyadic exchanges, individuals will often unconsciously imitate their partner. A substantial amount of research has been made on this phenomenon, and such studies have shown that synchronization between communication partners can improve interpersonal relationships. Automatic computational approaches for recognizing synchrony are still in their infancy. In this study, we extend on previous work in which we applied a novel method utilizing hand-crafted low-level acoustic descriptors and autoencoders (AEs) to analyse synchrony in the speech domain. For this purpose, a database consisting of 394 in-the-wild speakers from six different cultures, is used. For each speaker in the dyadic exchange, two AEs are implemented. Post the training phase, the acoustic features for one of the speakers is tested using the AE trained on their dyadic partner. In this same way, we also explore the benefits that deep representations from audio may have, implementing the state-of-the-art Deep Spectrum toolkit. For all speakers at varied time-points during their interaction, the calculation of reconstruction error from the AE trained on their respective dyadic partner is made. The results obtained from this acoustic analysis are then compared with the linguistic experiments based on word counts and word embeddings generated by our <em>word2vec</em> approach. The results demonstrate that there is a degree of synchrony during all interactions. We also find that, this degree varies across the 6 cultures found in the investigated database. These findings are further substantiated through the use of 4,096 dimensional Deep Spectrum features.</p>
Received funding from : Bavarian State Ministry of Education, Science and the Arts in the framework of the Centre Digitisation.Bavaria (ZD.B)
https://doi.org/10.3389/frobt.2019.00116
oai:zenodo.org:4294181
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
Frontiers in Robotics and AI, Front. Robot. AI, 08 November 2019(Computational Approaches for Human-Human and Human-Robot Social Interactions), https://www.frontiersin.org/articles/10.3389/frobt.2019.00116/, (2020-11-27)
speech synchronization
computational paralinguistics
human-human interaction
machine learning
speech processing
autoencoders
Synchronisation in Interpersonal Speech
info:eu-repo/semantics/article
oai:zenodo.org:6610232
2022-06-03T11:16:35Z
user-sustage
Grzegorz Sedek
Thomas Hess
Dayna Touron
2021-11-22
<p>Description</p>
<p>The study of aging and cognition has grown exponentially over the past 50 years, developing from a field dominated by experimentally based information-processing traditions to one represented by a more mature approach both conceptually and methodologically. In the past 10 years there has been growth in integrative approaches that incorporate behavioral, neuropsychological, and social information. In addition, there has been a growing recognition of the limitations associated with simple cross-sectional age-group comparisons, along with an increased use of more complex methods. This has resulted in the development of increasingly sophisticated research designs and analytic tools focused on understanding a multitude of potential mediators and moderators of cognitive change. The result has been a move away from negative-views of cognitive aging to one that is more nuanced and sensitive to contextual factors.<br>
<br>
<em>Multiple Pathways of Cognitive Aging</em> explores the factors associated with adaptive functioning in later life. Its emphasis is on understanding both the factors underlying individual differences in change in cognitive functioning in later life and the nature of the compensatory mechanisms developed by most successful and active middle-aged and older adults. This includes a consideration of motivational factors as a driver of both cognitive change and adaptive functioning.<br>
<br>
For students and researchers, <em>Multiple Pathways of Cognitive Aging </em>offers valuable insights into the field of cognitive development, along with innovative methodological approaches to help them in their own research.</p>
https://oxford.universitypressscholarship.com/view/10.1093/oso/9780197528976.001.0001/oso-9780197528976
https://doi.org/10.5281/zenodo.6610232
oai:zenodo.org:6610232
eng
Multiple Pathways of Cognitive Aging
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.6610231
info:eu-repo/semantics/restrictedAccess
cognitive aging
motivation
adaptive functioning
cognitive health
compensation
goals
decision making
engagement
Multiple Pathways of Cognitive Aging
info:eu-repo/semantics/bookPart
oai:zenodo.org:4294239
2020-11-28T12:27:16Z
user-sustage
user-eu
Meishu SONG
Adria MALLOL-RAGOLTA
Emilia PARADA-CABALEIRO
Zijiang YANG
Shuo LIU
Zhao REN
Ziping ZHAO
Björn W. SCHULLER
2020-08-28
<p>ABSTRACT</p>
<p><strong>Background</strong> Although frustration is a common emotional reaction during playing games, an excessive level of frustration can harm users’ experiences, discouraging them from undertaking further game interactions. The automatic detection of players’ frustration enables the development of adaptive systems, which through a real-time difficulty adjustment, would adapt the game to the user’s specific needs; thus, maximising players experience and guaranteeing the game success. To this end, we present our speech-based approach for the automatic detection of frustration during game interactions, a specific task still under-explored in research. <strong>Method </strong>The experiments were performed on the Multimodal Game Frustration Database (MGFD), an audiovisual dataset—collected within the Wizard-of-Oz framework—specially tailored to investigate verbal and facial expressions of frustration during game interactions. We explored the performance of a variety of acoustic feature sets, including Mel-Spectrograms and Mel-Frequency Cepstral Coefficients (MFCCs), as well as the low dimensional knowledge-based acoustic feature set eGeMAPS. Due to the always increasing improvements achieved by the use of Convolutional Neural Networks (CNNs) in speech recognition tasks, unlike the MGFD baseline—based on Long Short-Term Memory (LSTM) architecture and Support Vector Machine (SVM) classifier—in the present work we take into consideration typically used CNNs, including ResNets, VGG, and AlexNet. Furthermore, given the still open debate on the shallow vs deep networks suitability, we also examine the performance of two of the latest deep CNNs, i. e., WideResNets and EfficientNet. <strong>Results </strong>Our best result, achieved with WideResNets and Mel-Spectrogram features, increases the system performance from 58.8 % Unweighted Average Recall (UAR) to 93.1 % UAR for speech-based automatic frustration recognition.</p>
https://doi.org/10.3724/SP.J.2096-5796.20.00090
oai:zenodo.org:4294239
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
VRIH 2020, Virtual Reality & Intelligent Hardware
Frustration Recognition
WideResNets
Machine Learning
Frustration recognition from speech during game interaction using wide residual networks.
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4304867
2020-12-09T12:26:58Z
user-sustage
user-eu
Adria Mallol-Ragolta
Shuo Liu
Nicholas Cummins
Björn Schuller
2020-12-01
<p>The high prevalence of chronic pain in society raises the need to develop new digital tools that can automatically and objectively assess pain intensity in individuals. These tools can contribute to an optimisation of clinical resources, as they offer cost-effective solutions for early detection, continuous monitoring, and treatment personalisation by utilising Artificial Intelligence techniques. In this work, we present our contribution to the Pain Intensity Estimation from Facial Expressions task of the EMOPAIN 2020 Challenge. Specifically, we compare the performance of Recurrent Neural Networks trained with standard or Curriculum Learning (CL) approaches to predict the pain intensity level of individuals reported in an 11-point scale from facial expressions. The results obtained using the test partition support the use of CL-based approaches in the automatic prediction of pain from facial features. The best model trained using a CL approach achieved a Concordance Correlation Coefficient (CCC) of 0.196 in the test partition, while the model trained using a standard approach, without CL, achieved a CCC of 0.174. In terms of CCC, these results respectively represent an improvement of 0.136 and 0.114 on the best results of the baseline system reported by the Challenge organisers using the test partition.</p>
FUNDING: Spanish Ministry of Economy and Competitiveness under project grant TIN2017-90124-P, the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), and the donation bahi2018-19to the CMTech at UPF.
https://doi.org/10.5281/zenodo.4304867
oai:zenodo.org:4304867
eng
IEEE
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.4304866
info:eu-repo/semantics/restrictedAccess
FG 2020, 15th IEEE International Conference on Automatic Face and Gesture Recognition, Buenos Aires, AR, 16 - 20 May 20 2020 ISBN:
A Curriculum Learning Approach for Pain Intensity Recognition from Facial Expressions
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6651976
2022-06-16T11:32:32Z
user-sustage
user-eu
Shuo Liu
Adria Mallol-Ragolta
Tianhao Yan
Kun Qian
Emilia Parada-Cabaleiro
Bin Hu
Bjorn Schuller
2022-05-06
<p><strong>Abstract:</strong></p>
<p>The importance of detecting whether a per- son wears a face mask while speaking has tremendously increased since the outbreak of SARS-CoV-2 (COVID-19), as wearing a mask can help to reduce the spread of the virus and mitigate the public health crisis. Besides affecting human speech characteristics related to frequency, face masks cause temporal interferences in speech, altering the pace, rhythm, and pronunciation speed. In this regard, this paper presents two effective neural network models to detect surgical masks from audio. The proposed architectures are both based on Convolutional Neural Net- works (CNNs), chosen as an optimal approach for the spatial processing of the audio signals. One architecture applies a Long Short-Term Memory (LSTM) network to model the time-dependencies. Through an additional attention mechanism, the LSTM-based architecture enables the extraction of more salient temporal information. The other architecture (named ConvTx) retrieves the relative position of a sequence through the positional encoder of a transformer module. In order to assess to which extent both architectures can complement each other when modelling temporal dynamics, we also explore the combination of LSTM and Transformers in three hybrid models. Finally, we also investigate whether data augmentation techniques, such as, using transitions between audio frames and considering gender-dependent frameworks might impact the performance of the proposed architectures. Our experimental results show that one of the hybrid models achieves the best performance, surpassing existing state-of-the-art results for the task at hand.</p>
https://doi.org/10.1109/JBHI.2022.3173128
oai:zenodo.org:6651976
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/restrictedAccess
IEEE Journal of Biomedical and Health Informatics, (2022-05-06)
Face recognition
Spectrogram
Neural networks
Task analysis
Feature extraction
COVID-19
Acoustics
Capturing Time Dynamics from Speech using Neural Networks for Surgical Masks Detection
info:eu-repo/semantics/article
oai:zenodo.org:6610346
2022-06-03T09:53:03Z
user-sustage
user-eu
Shuo Liu
Adria Mallol-Ragolta
Björn W. Schuller
2021-12-09
<p><strong>Abstract:</strong></p>
<p>This study explores the use of deep learning-based methods for the automatic detection of COVID-19. Specifically, we aim to investigate the involvement of the virus in the respiratory system by analysing breathing and coughing sounds. Our hypothesis resides in the complementarity of both data types for the task at hand. Therefore, we focus on the analysis of fusion mechanisms to enrich the information available for the diagnosis. In this work, we introduce a novel injection fusion mechanism that considers the embedded representations learned from one data type to extract the embedded representations of the other data type. Our experiments are performed on a crowdsourced database with breathing and coughing sounds recorded using both a web-based application, and a smartphone app. The results obtained support the feasibility of the injection fusion mechanism presented, as the models trained with this mechanism outperform single-type models and multi-type models using conventional fusion mechanisms.</p>
https://doi.org/10.5281/zenodo.6610346
oai:zenodo.org:6610346
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610345
info:eu-repo/semantics/restrictedAccess
EMBC, 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society, Mexico, 01-05 November 2021
COVID-19
Learning systems
Technological innovation
Fuses
Databases
Europe
Feature extraction
COVID-19 Detection with a Novel Multi-Type Deep Fusion Method using Breathing and Coughing Information
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:7092354
2022-09-20T02:26:28Z
user-sustage
user-eu
Adria Mallol - Ragolta
Michail Maniadakis
George Papadopoulos
Bjoern Schuller
2022-09-19
<p>Abstract: This work investigates the use of TS2Vec time series representations in an end-to-end approach to detect the fatigue levels perceived by workers of the transportation and logistics industry from the analysis of the accelerometer and the heart rate measurements sensed using a Garmin Vivoactive 3 device. The experiments are conducted using the dataset collected during a pre-pilot study with a total of 1 h 22 min 20 sec of data available. The results obtained support the use of TS2Vec representations for the task at hand, as the binary model trained using this approach and exploiting the heart rate modality obtains the best performance with an Unweighted Average Recall of 67.1 %.</p>
https://doi.org/10.54941/ahfe1002147
oai:zenodo.org:7092354
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
AHFE 2022, International Conference on Applied Human Factors and Ergonomics, New York, USA, 24-28 July 2022
Fatigue estimation
AI - Artificial intelligence
ubiquitous sensing
Time series representation
Wearable devices
Time Series Representation using TS2Vec on Smartwatch Sensor Data for Fatigue Estimation
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6652038
2022-11-18T14:41:09Z
user-sustage
user-felice-h2020
user-eu
Konstantinos Papoutsakis
George Papadopoulos
Michail Maniadakis
Thodoris Papadopoulos
Manolis Lourakis
Maria Pateraki
Iraklis Varlamis
2022-03-16
<p>Abstract</p>
<p>The detection and prevention of workers’ body straining postures and other stressing conditions within the work environment, supports establishing occupational safety and promoting well being and sustainability at work. Developed methods towards this aim typically rely on combining highly ergonomic workplaces and expensive monitoring mechanisms including wearable devices. In this work, we demonstrate how the input from low-cost sensors, specifically, passive camera sensors installed in a real manufacturing workplace, and smartwatches used by the workers can provide useful feedback on the workers’ conditions and can yield key indicators for the prevention of work-related musculo-skeletal disorders (WMSD) and physical fatigue. To this end, we study the ability to assess the risk for physical strain of workers online during work activities based on the classification of ergonomically sub-optimal working postures using visual information, the correlation and fusion of these estimations with synchronous worker heart rate data, as well as the prediction of near-future heart rate using deep learning-based techniques. Moreover, a new multi-modal dataset of video and heart rate data captured in a real manufacturing workplace during car door assembly activities is introduced. The experimental results show the efficiency of the proposed approach that exceeds 70% of classification rate based on the F1 score measure using a set of over 300 annotated video clips of real line workers during work activities. In addition a time lagging correlation between the estimated ergonomic risks for physical strain and high heart rate was assessed using a larger dataset of synchronous visual and heart rate data sequences. The statistical analysis revealed that imposing increased strain to body parts will results in an increase to the heart rate after 100–120 s. This finding is used to improve the short term forecasting of worker’s cardiovascular activity for the next 10 to 30 s by fusing the heart rate data with the estimated ergonomic risks for physical strain and ultimately to train better predictive models for worker fatigue</p>
https://doi.org/10.3390/technologies10020042
oai:zenodo.org:6652038
eng
Zenodo
https://zenodo.org/communities/felice-h2020
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
Technologies 2022, (2022-03-16)
computer vision
sensor fusion
low cost sensors
heart rate
WMSD
fatigue
ergonomic risk
physical strain
working postures
predictive models
occupational health
Detection of Physical Strain and Fatigue in Industrial Environments Using Visual and Non-Visual Low-Cost Sensors
info:eu-repo/semantics/article
oai:zenodo.org:7043787
2022-11-11T14:50:26Z
openaire_data
user-sustage
user-felice-h2020
user-eu
Konstantinos Papoutsakis
Georgios Papadopoulos
Michail Maniadakis
Thodoris Papadopoulos
Manolis Lourakis
Maria Pateraki
Iraklis Varlamis
2022-09-02
<p>The sustAGE User Postures & Actions Monitoring dataset contains videos, data related to occurrences of ergonomic (body straining) postures as well as heart rate measurements of line workers during work activities in a realistic manufacturing environment. The time-synchronized data streams were captured during assembly tasks performed by real line workers in a realistic car door assembly production line in the premises of Stellantis --- Centro Ricerche FIAT (CRF)/SPW Research \& Innovation department in Melfi, Italy in the context of the sustAGE project.</p>
<p>The dataset can be used by methods that target the following research tasks: a) vision-based detection of worker’s ergonomically unhealthy (body straining) postures during assembly tasks according to the MURI risk analysis tool, b) vision-based recognition of human assembly actions, c) multi-modal analysis and forecasting of worker heart rate and physical fatigue based on heart rate measurements and the detection (annotated occurrences) of body straining postures during work activities. Each video shows a single line worker that performs a series of car door assembly activities, noted as a task cycle execution, for a specific workstation of the production line. A set of twelve task cycle executions performed by three different workers were captured using static StereoLabs ZED or ZED2 sensors. Each task cycle execution has a mean duration of 4 minutes. Moreover, time-synchronised data for the occurrences of body straining postures and worker heart rate are available for 8 work sessions. Each work session was recorded during morning or afternoon time of the work shift and regards 3 to 5 consecutive task cycle executions performed by the same line worker. Heart rate data was captured using a Garmin Vivoactive 3 smartwatch. Time annotations are available for the samples collected during the work sessions based on the Unix time (Epoch time) format.</p>
<p>Annotation data are available for twelve task cycle executions and contain the semantic labels and temporal boundaries of each assembly action and for the ergonomic risk scores of body straining postures performed by the line worker during each task cycle execution. Data was provided by experts in manufacturing and ergonomics based on the MURI risk analysis tool.</p>
<p>Documentation of this dataset can be found in [1].</p>
<p>If you use the dataset in your research work, you are kindly asked to cite [1] in your publications.</p>
<p>[1] Papoutsakis K, Papadopoulos G, Maniadakis M, Papadopoulos T, Lourakis M, Pateraki M, Varlamis I. Detection of Physical Strain and Fatigue in Industrial Environments Using Visual and Non-Visual Low-Cost Sensors. MDPI <em>Technologies</em>. 2022; 10(2):42. <a href="https://doi.org/10.3390/technologies10020042">https://doi.org/10.3390/technologies10020042</a></p>
https://doi.org/10.5281/zenodo.7043787
oai:zenodo.org:7043787
Zenodo
https://zenodo.org/communities/felice-h2020
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.7043786
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
MDPI Technologies - Collection Selected Papers from the PETRA Conference Series, 10(2), 42, (2022-09-02)
ergonomic body postures
visual recognition
heart rate analysis
physical fatigue
dataset
computer vision
manufacturing activities
sustAGE User Postures & Actions Monitoring dataset
info:eu-repo/semantics/other
oai:zenodo.org:6610388
2022-06-03T10:09:11Z
user-sustage
user-eu
Manolis Lourakis
Maria Pateraki
2021-11-24
<p><strong>Abstract:</strong></p>
<p>Crane systems play a crucial role in container transport logistics. This paper presents an approach for visually tracking the position and orientation in 3D space of a container crane spreader. An initial pose estimate is first employed to render a 3D triangle mesh model of the spreader as a wireframe with hidden lines removed. The initial pose is then refined so that the visible lines of the wireframe match the straight line segments detected in an input image. Line segment matching relies on fast, local one-dimensional searches along a segment’s normal direction. Matched line segments yield constraints on the spreader motion which are processed with robust parameter estimation techniques that safeguard against outliers stemming from mismatches. The tracker automatically determines the visibility of segments, without making limiting assumptions regarding the spreader’s 3D mesh model. It is also robust to parts of the tracked spreader being out of view, occluded, shadowed or simply undetected. Experimental results demonstrating the tracker’s performance are additionally included.</p>
https://doi.org/10.5281/zenodo.6610388
oai:zenodo.org:6610388
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610387
info:eu-repo/semantics/restrictedAccess
ICCVW, 2021 IEEE/CVF International Conference on Computer Vision Workshops (), Montreal, BC, Canada, 11-17 October 2021
Image segmentation
Solid modeling
Visualization
Cranes
Three-dimensional displays
Tracking
Motion segmentation
Markerless Visual Tracking of a Container Crane Spreader
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4312834
2020-12-17T00:27:07Z
user-sustage
user-eu
George Terzakis
Manolis Lourakis
2020-11-03
<p>ABSTRACT:</p>
<p>An approach for estimating the pose of a camera given a set of 3D points and their corresponding 2D image projections is presented. It formulates the problem as a non-linear quadratic program and identifies regions in the parameter space that contain unique minima with guarantees that at least one of them will be the global minimum. Each regional minimum is computed with a sequential quadratic programming scheme. These premises result in an algorithm that always determines the global minima of the perspective-n-point problem for any number of input correspondences, regardless of possible coplanar arrangements of the imaged 3D points. For its implementation, the algorithm merely requires ordinary operations available in any standard off-the-shelf linear algebra library. Comparative evaluation demonstrates that the algorithm achieves state-of-the-art results at a consistently low computational cost.</p>
pp 478-494
M. Lourakis has been funded by the EU H2020 Programme under Grant Agreement No. 826506 (sustAGE).
https://doi.org/10.1007/978-3-030-58452-8_28
oai:zenodo.org:4312834
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
Computer Vision – ECCV 2020, 16th European Conference on Computer Vision, Glasgow, UK, August 23–28 2020
Perspective-n-point problem
Non-linear quadratic program
Sequential quadratic programming
Global optimality
Pose estimation
A Consistently Fast and Globally Optimal Solution to the Perspective-n-Point Problem
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6610302
2022-06-03T13:50:44Z
user-sustage
Jennifer A. Rieker
José Manuel Reales
Soledad Ballesteros
2020-12-18
<p>Findings suggest a positive impact of bilingualism on cognition, including the later onset of dementia. However, it is not clear to what extent these effects are influenced by variations in attentional control demands in response to specific task requirements. In this study, 20 bilingual and 20 monolingual older adults performed a task-switching task under explicit task-cuing vs. memory-based switching conditions. In the cued condition, task switches occurred in random order and a visual cue signaled the next task to be performed. In the memory-based condition, the task alternated after every second trial in a predictable sequence without presenting a cue. The performance of bilinguals did not vary across experimental conditions, whereas monolinguals experienced a pronounced increase in response latencies and error rates in the cued condition. Both groups produced similar switch costs (difference in performance on switch trials as opposed to repeating trials within the mixed-task block) and mixing costs (difference in performance on repeat trials of a mixed-task block as opposed to trials of a single-task block), but bilinguals produced them with lower response latencies. The cognitive benefits of bilingualism seem not to apply to executive functions <em>per se</em> but to affect specific cognitive processes that involve task-relevant context processing. The present results suggest that lifelong bilingualism could promote in older adults a flexible adjustment to environmental cues, but only with increased task demands. However, due to the small sample size, the results should be interpreted with caution.</p>
https://www.frontiersin.org/articles/10.3389/fnhum.2020.610548/full
https://doi.org/10.5281/zenodo.6610302
oai:zenodo.org:6610302
eng
Zenodo
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.6610301
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
Frontiers in in Human Neuroscience, Cognitive Neuroscience, 18 December 2020, (2020-12-18)
The Effect of Bilingualism on Cue-Based vs. Memory-Based Task Switching in Older Adults
info:eu-repo/semantics/article
oai:zenodo.org:4106489
2020-11-28T12:27:15Z
user-sustage
Soledad Ballesteros
Jennifer A Rieker
Julia Mayas
Antonio Prieto
Pilar Toril
María Pilar Jiménez 3
José Manuel Reales 3
2020-05-14
<p><strong>Background: </strong> Previous research suggests that both cognitive training and physical exercise help to maintain brain health and cognitive functions that decline with age. Some studies indicate that combined interventions may produce larger effects than each intervention alone. The aim of this study is to investigate the effects of combined cognitive and physical training compared to cognitive training and physical training alone on executive control and memory functions in healthy older adults.</p>
<p><strong>Objectives: </strong> The main objectives of this four-arm randomized controlled trial (RCT) are: to investigate the synergetic effects of a simultaneous, group-based multidomain training program that combines cognitive video-game training with physical exercise, in comparison to those produced by cognitive training combined with physical control activity, physical training combined with cognitive control activity, or a combination of both control activities; to investigate whether event-related potential latencies of the P2 component are shorter and N2 and P3b components assessed in a memory-based task switching task are enhanced after training; and to find out whether possible enhancements persist after a 3-month period without training.</p>
<p><strong>Methods: </strong> In this randomized, single-blind, controlled trial, 144 participants will be randomly assigned to one of the four combinations of cognitive training and physical exercise. The cognitive component will be either video-game training (cognitive intervention, CI) or video games not specifically designed to train cognition (cognitive control, CC). The physical exercise component will either emphasize endurance, strength, and music-movement coordination (exercise intervention, EI) or stretching, toning, and relaxation (exercise control, EC).</p>
<p><strong>Discussion: </strong> This RCT will investigate the short and long-term effects of multidomain training, compared to cognitive training and physical training alone, on executive control and memory functions in healthy older adults, in comparison with the performance of an active control group.</p>
<p><strong>Trial registration: </strong> ClinicalTrials.gov, <a href="http://clinicaltrials.gov/show/NCT03823183">NCT03823183</a>. Registered on 21 January 2019.</p>
This clinical trial was registered at the National Institute of Health (NIH) with the Clinicaltrials.gov identifier NCT03823183 (https://register.clinicaltrials.gov/ClinicalTrials.gov) on 21 January 2019. The protocol version number is number 1 (January 2019). Recruitment started in February 2019 and is expected to be completed in February 2020. Once the trial is completed, results will be reported according to the Consolidated Standards of Reporting Trials (CONSORT) guidelines. The trial is active and ongoing. We expect to have the final results by the middle of 2021.
https://doi.org/10.1186/s13063-020-04293-3
oai:zenodo.org:4106489
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
BMC Part of Springer Nature, Study Protocol, TRIALS, TRIALS, Article number: 404 (2020), (2020-05-14)
PMID: 32410715, BioMed Central
Aging; Cognitive training; Executive functions; Memory functions; Multidomain training; Physical exercise; Randomized controlled trial.
Effects of multidomain versus single-domain training on executive control and memory in older adults: study protocol for a randomized controlled trial
info:eu-repo/semantics/article
oai:zenodo.org:4349556
2022-06-03T09:45:13Z
user-sustage
JENNIFER RIEKER
JOSÉ REALES
MÓNICA MUIÑOS
SOLEDAD BALLESTEROS
2020-11-22
<p>The effectiveness of multidomain training compared to cognitive or physical training alone has been controversial with some studies suggesting that combined interventions might produce synergetic effects. We conducted a three-level meta-analysis on the transfer effects of multidomain interventions versus cognitive and physical training alone. We obtained 1,070 effect sizes from 54 studies, involving 5.547 healthy older adults. Our results revealed a synergetic effect of multidomain training on executive functions, and larger effects on attention and memory than cognitive and physical training. Multidomain and single cognitive training produced similar effects on memory in comparison to physical training. We did not find differences in processing speed, verbal functions, and global cognition. Moderator analyses showed a complex pattern. In general, age, publication year, and study quality were not significant. We conclude that the combination of cognitive training with physical exercise could be a promising strategy to prevent cognitive and physical declines with aging.</p>
https://doi.org/10.5281/zenodo.4349556
oai:zenodo.org:4349556
eng
Zenodo
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.4349555
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
Psychonomic Society's 2020 Annual Meeting, Virtual, 19-22 November 2020
Multidomain Training in Healthy Older Adults Revisited: A Three-Level Meta-Analysis
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4294231
2020-11-28T12:27:15Z
user-sustage
user-eu
Lukas Stappen
Alice Baird
Georgios Rizos
Panagiotis Tzirakis
Xinchen Du
Felix Hafner
Lea Schumann
Adria Mallol-Ragolta
Björn Schuller
Iulia Lefter
Erik Cambr
2020-11-27
<p>ABSTRACT</p>
<p>Multimodal Sentiment Analysis in Real-life Media (MuSe) 2020 is a Challenge-based Workshop focusing on the tasks of sentiment recognition, as well as emotion-target engagement and trustworthiness detection by means of more comprehensively integrating the audio-visual and language modalities. The purpose of MuSe 2020 is to bring together communities from different disciplines; mainly, the audio-visual emotion recognition community (signal-based), and the sentiment analysis community (symbol-based). We present three distinct sub-challenges: MuSe-Wild, which focuses on continuous emotion (arousal and valence) prediction; MuSe-Topic, in which participants recognise 10 domain-specific topics as the target of 3-class (low, medium, high) emotions; and MuSe-Trust, in which the novel aspect of trustworthiness is to be predicted. In this paper, we provide detailed information on MuSe-CAR, the first of its kind in-the-wild database, which is utilised for the challenge, as well as the state-of-the-art features and modelling approaches applied. For each sub-challenge, a competitive baseline for participants is set; namely, on test we report for MuSe-Wild a combined (valence and arousal) CCC of .2568, for MuSe-Topic a score (computed as 0.34 * UAR + 0.66 * F1) of 76.78 % on the 10-class topic and 40.64 % on the 3-class emotion prediction, and for MuSe-Trust a CCC of .4359.</p>
Funding from the EP- SRC Grant No. 2021037, and the Bavarian State Ministry of Education, Science and the Arts in the framework of the Centre Digitisation.Bavaria (ZD.B). We thank the sponsors of the Challenge BMW Group and audEERING.
https://doi.org/10.1145/3423327.3423673
oai:zenodo.org:4294231
eng
Association for Computing Machinery
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
MuSe '20, MuSe'20: Proceedings of the 1st International on Multimodal Sentiment Analysis in Real-life Media Challenge and Workshop, Seattle, WA, USA, October 16, 2020
Multimodal Sentiment Analysis
Affective Computing
User- Generated Data
Multimodal Fusion
MuSe 2020 Challenge and Workshop: Multimodal Sentiment Analysis, Emotion-target Engagement and Trustworthiness Detection in Real-life Media: Emotional Car Reviews in-the-wild
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:3610149
2020-11-27T23:29:59Z
user-sustage
user-eu
Sotiris Nousias
Manolis Lourakis
Christos Bergeles
2019-06-15
<p>This paper presents a large scale, metric Structure from Motion (SfM) pipeline for generalised cameras with overlapping fields-of-view, and demonstrates it using Light Field (LF) images. We build on recent developments in algorithms for absolute and relative pose recovery for generalised cameras and couple them with multi-view triangulation in a robust framework that advances the state-of-the-art on 3D reconstruction from LFs in several ways. First, our framework can recover the scale of a scene. Second, it is concerned with unordered sets of LF images, meticulously determining the order in which images should be considered. Third, it can scale to datasets with hundreds of LF images. Finally, it recovers 3D scene structure while abstaining from triangulating using very small baselines. Our approach outperforms the state-of-the-art, as demonstrated by real-world experiments with variable size datasets.</p>
https://doi.org/10.1109/CVPR.2019.00341
oai:zenodo.org:3610149
eng
Zenodo
http://openaccess.thecvf.com/content_CVPR_2019/html/Nousias_Large-Scale_Metric_Structure_From_Motion_for_Unordered_Light_Fields_CVPR_2019_paper.html
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
CVPR, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA, 15-20 June 2019
cameras
image motion analysis
image reconstruction
3D from Multiview and Sensors
Large-Scale, Metric Structure From Motion for Unordered Light Fields
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4294274
2020-11-28T12:27:15Z
user-sustage
Soledad Ballesteros
Jennifer Rieker
Jjosé M Reales
Julia Mayas
María Pilar Jiménez
Antonio Prieto
Pilar Toril
2019-11-08
<p>Abstract</p>
<p>Previous research suggests that both cognitive training and physical exercise help to maintain brain health and cognitive functions that decline with age. The main objectives of this four-arms RCT are (1) to investigate the synergetic effects of a group-based multidomain training program that combines cognitive video-game training with physical exercise, in comparison to those produced by cognitive training combined with physical control activity, physical training combined with cognitive control activity, or a combination of both control activities; (2) to investigate in a memory-based task switching task whether event Related Potential (ERP) latencies of the P2 component are shorter, and N2 and P3b components are enhanced after training; and (3) to find out whether possible enhancements persist after a 3-month period without training. One hundred and twenty participants will be randomly assigned to one of the four combinations of cognitive training and physical exercise. The cognitive component will be either video-game training (cognitive intervention, CI) or video games not specifically designed to train cognition (cognitive control, CC). The physical exercise component will either emphasize endurance, strength, and music-movement coordination (exercise intervention, EI) or stretching, toning and relaxation (exercise control, EC). This RCT will investigate the short and long-term effects of combined multi-domain training compared to cognitive training and physical training alone, on executive control and memory functions of healthy older adults, in comparison with the performance of an active control group. This trial is an ongoing project started in 2018. Trial registration: Clinicaltrials.gov ID: NCT03823183; <a href="https://register.clinicaltrials.gov/">https://register.clinicaltrials.gov/</a></p>
https://doi.org/10.1093/geroni/igz038.2644
oai:zenodo.org:4294274
eng
Oxford University Press
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
Innovation in Aging, Volume 3(Issue Supplement_1 - Section SUCCESSFUL AGING), Page S721, (2019-11-08)
Is combined training more effective thatn single-domain training: A randomized controlled trial with older adults
info:eu-repo/semantics/article
oai:zenodo.org:4294199
2020-11-28T12:27:15Z
user-sustage
user-eu
Fabien Ringeval
Björn Schuller
Michel Valstar
Nicholas Cummins
Roddy Cowie
Leili Tavabi
Maximilian Schmitt
Sina Alisamir
Shahin Amiriparian
Eva-Maria Messner
Siyang Song
Shuo Liu
Ziping Zhao
Adria Mallol-Ragolta
Zhao Ren
Mohammad Soleymani
Maja Pantic
2019-10-15
<p>ABSTRACT</p>
<p>The Audio/Visual Emotion Challenge and Workshop (AVEC 2019) 'State-of-Mind, Detecting Depression with AI, and Cross-cultural Affect Recognition' is the ninth competition event aimed at the comparison of multimedia processing and machine learning methods for automatic audiovisual health and emotion analysis, with all participants competing strictly under the same conditions. The goal of the Challenge is to provide a common benchmark test set for multimodal information processing and to bring together the health and emotion recognition communities, as well as the audiovisual processing communities, to compare the relative merits of various approaches to health and emotion recognition from real-life data. This paper presents the major novelties introduced this year, the challenge guidelines, the data used, and the performance of the baseline systems on the three proposed tasks: state-of-mind recognition, depression assessment with AI, and cross-cultural affect sensing, respectively.</p>
Further funding has also been received from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement No. 115902, which receives support from the European Union's Horizon 2020 research and innovation program and EFPIA. The work on the DDS was supported in part by the U.S. Army. Any opinion, content or information presented does not necessarily re ect the position or the policy of the United States Government, and no o cial endorse- ment should be inferred. The authors further thank the sponsor of the challenge – audEERING GmbH.
https://doi.org/10.1145/3347320.3357688
oai:zenodo.org:4294199
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
ACM Multimedia 2019, AVEC 2019-Proceedings of the 9th International Audio/Visual Emotion Challenge and Workshop, 15-October 2019
Affective Computing
State-of-Mind
Cross-Cultural Emotion
AVEC 2019 Workshop and Challenge: State-of-Mind, Detecting Depression with AI, and Cross-Cultural Affect Recognition
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6610577
2022-06-03T11:14:57Z
user-sustage
G Athanassiou
P Gajewski
A Ascolese
S Ballesteros
M Maniadakis
M Pateraki
A Prieto
I Varlamis
R Monferino
2021-09-22
<p>Purpose Cognitive ageing may impair the ability of older employees to remain in the workforce. The interdisciplinary EU-project sustAGE draws upon digital technology trends to support older workers by providing guidance for health-promoting activities in and outside the working context. One of sustAGE’s objectives is the enhancement of age-affected cognitive abilities through targeted training interventions. A combination of personalized recommendations and corresponding cognitive trainings in form of digital serious games serves this purpose. Three interlinked steps are necessary for implementing the sustAGE approach: base-line assessment of cognitive abilities of potential groups of users; derivation and implementation of user-specific recommendations; and development and evaluation of training. The first step within sustAGE development included an empirical assessment of cognitive abilities from a sample of older employees in two of the most important sectors of EU industry: automotive industry (AI) and maritime logistics (ML).</p>
<p>Methods A psychometric test battery measuring attentional, memory and executive functions was administered to a sample of 60 older employees (M = 53.4; SD = 5.1) from the two industries.</p>
<p>Results Comparisons between occupational groups revealed significant effects regarding cognitive performance (AI > HL; sustained attention, p < .0001; processing speed, p < .005; task switching, p < .0001).</p>
<p>Conclusions Group differences in cognitive performance may relate to specific job characteristics of the two occupational domains and provide evidence for differentiated, group-specific needs and respective interventions. Current results as well as additional sample analysis will serve as basis for the development of targeted recommendations and cognitive trainings based on the inferred group profiles.</p>
https://doi.org/10.5281/zenodo.6610577
oai:zenodo.org:6610577
eng
Zenodo
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.6610576
info:eu-repo/semantics/restrictedAccess
Sustainable work through technology-assisted enhancement of cognitive abilities of older employees: the sustAGE approach.
info:eu-repo/semantics/article
oai:zenodo.org:5801362
2021-12-27T10:04:50Z
user-sustage
Mostafa M.Mohamed
Mina A.Nessiem
Anton Batliner
Christian Bergler
Simone Hantkee
Maximilian Schmitt
Alice Baird
Adria Mallol-Ragoltaa
Vincent Karas
Shahin Amiriparian
2021-10-04
<p>Abstract</p>
<p>The sudden outbreak of COVID-19 has resulted in tough challenges for the field of <a href="https://www.sciencedirect.com/topics/engineering/biometric">biometrics</a> due to its spread via physical contact, and the regulations of wearing face masks. Given these constraints, voice biometrics can offer a suitable contact-less biometric solution; they can benefit from models that classify whether a speaker is wearing a mask or not. This article reviews the Mask Sub-Challenge (MSC) of the INTERSPEECH 2020 COMputational PARalinguistics challengE (ComParE), which focused on the following <a href="https://www.sciencedirect.com/topics/engineering/classification-task">classification task</a>: Given an audio chunk of a speaker, classify whether the speaker is wearing a mask or not. First, we report the collection of the Mask Augsburg Speech Corpus (MASC) and the baseline approaches used to solve the problem, achieving a performance of Unweighted Average Recall (UAR). We then summarise the methodologies explored in the submitted and accepted papers that mainly used two common patterns: (i) phonetic-based audio features, or (ii) <a href="https://www.sciencedirect.com/topics/engineering/spectrogram">spectrogram</a> representations of audio combined with <a href="https://www.sciencedirect.com/topics/engineering/convolutional-neural-networks">Convolutional Neural Networks</a> (CNNs) typically used in <a href="https://www.sciencedirect.com/topics/engineering/image-processing">image processing</a>. Most approaches enhance their models by adapting ensembles of different models and attempting to increase the size of the training data using various techniques. We review and discuss the results of the participants of this sub-challenge, where the winner scored a UAR of . Moreover, we present the results of fusing the approaches, leading to a UAR of . Finally, we present a smartphone app that can be used as a proof of concept demonstration to detect in real-time whether users are wearing a face mask; we also benchmark the run-time of the best models.</p>
https://doi.org/10.1016/j.patcog.2021.108361
oai:zenodo.org:5801362
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/restrictedAccess
Pattern Recognition ELSEVIER, 122(108361), (2021-10-04)
COVID-19
Deep learning
Masks
Voice biometrics
Acoustic modelling
Face mask recognition from audio: The MASC database and an overview on the mask challenge
info:eu-repo/semantics/article
oai:zenodo.org:6610339
2022-06-03T13:50:42Z
user-sustage
JENNIFER RIEKER
JOSÉ REALES
MÓNICA MUIÑOS
SOLEDAD BALLESTEROS
2020-11-22
<p>The effectiveness of multidomain training compared to cognitive or physical training alone has been controversial with some studies suggesting that combined interventions might produce synergetic effects. We conducted a three-level meta-analysis on the transfer effects of multidomain interventions versus cognitive and physical training alone. We obtained 1,070 effect sizes from 54 studies, involving 5.547 healthy older adults. Our results revealed a synergetic effect of multidomain training on executive functions, and larger effects on attention and memory than cognitive and physical training. Multidomain and single cognitive training produced similar effects on memory in comparison to physical training. We did not find differences in processing speed, verbal functions, and global cognition. Moderator analyses showed a complex pattern. In general, age, publication year, and study quality were not significant. We conclude that the combination of cognitive training with physical exercise could be a promising strategy to prevent cognitive and physical declines with aging.</p>
https://doi.org/10.5281/zenodo.6610339
oai:zenodo.org:6610339
eng
Zenodo
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.4349555
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
61st Psychonomic Society's 2020 Annual Meeting, Virtual, 19-22 November 2020
Multidomain Training in Healthy Older Adults Revisited: A Three-Level Meta-Analysis
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4349655
2022-06-03T09:45:14Z
user-sustage
JENNIFER RIEKER
JOSÉ REALES
MÓNICA MUIÑOS
SOLEDAD BALLESTEROS
2020-11-22
<p>The effectiveness of multidomain training compared to cognitive or physical training alone has been controversial with some studies suggesting that combined interventions might produce synergetic effects. We conducted a three-level meta-analysis on the transfer effects of multidomain interventions versus cognitive and physical training alone. We obtained 1,070 effect sizes from 54 studies, involving 5.547 healthy older adults. Our results revealed a synergetic effect of multidomain training on executive functions, and larger effects on attention and memory than cognitive and physical training. Multidomain and single cognitive training produced similar effects on memory in comparison to physical training. We did not find differences in processing speed, verbal functions, and global cognition. Moderator analyses showed a complex pattern. In general, age, publication year, and study quality were not significant. We conclude that the combination of cognitive training with physical exercise could be a promising strategy to prevent cognitive and physical declines with aging.</p>
https://doi.org/10.5281/zenodo.4349655
oai:zenodo.org:4349655
eng
Zenodo
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.4349555
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
61st Psychonomic Society's 2020 Annual Meeting, Virtual, 19-22 November 2020
Multidomain Training in Healthy Older Adults Revisited: A Three-Level Meta-Analysis
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:7024715
2022-08-28T02:26:20Z
user-sustage
Vincent Karas
Mani Kumar Tellamekala
Adria Mallol-Ragolta
Michel Valstar
Björn W. Schuller
2022-03-29
<p>In this paper, we present our submission to 3rd Affective Behavior Analysis in-the-wild (ABAW) challenge. Learningcomplex interactions among multimodal sequences is critical to recognise dimensional affect from in-the-wild audiovisual data. Recurrence and attention are the two widely used sequence modelling mechanisms in the literature. To clearly understand the performance differences between recurrent and attention models in audiovisual affect recognition, we present a comprehensive evaluation of fusion models based on LSTM-RNNs, self-attention and cross-modal attention, trained for valence and arousal estimation. Particularly, we study the impact of some key design choices: the modelling complexity of CNN backbones that provide features to the the temporal models, with and without end-to-end learning. We trained the audiovisual affect recognition models on in-the-wild ABAW corpus by systematically tuning the hyper-parameters involved in the network architecture design and training optimisation. Our extensive evaluation of the audiovisual fusion models shows that LSTM-RNNs can outperform the attention models when coupled with low-complex CNN backbones and trained in an end-to-end fashion, implying that attention models may not necessarily be the optimal choice for continuous-time multimodal emotion recognition.</p>
https://doi.org/10.48550/arXiv.2203.13285
oai:zenodo.org:7024715
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
ArXiv Computer Science Sound, (2022-03-29)
Continuous-Time Audiovisual Fusion with Recurrence vs. Attention for In-The-Wild Affect Recognition
info:eu-repo/semantics/article
oai:zenodo.org:4294256
2020-12-09T12:16:53Z
user-sustage
user-eu
Maria Pateraki
Manolis Lourakis
Leonidas Kallipolitis
Frank Werner
Petros Patias
Christos Pikridas
2019-10-08
<p><a href="https://ercim-news.ercim.eu/en119/special/supporting-the-wellness-at-work-and-productivity-of-ageing-employees-in-industrial-environments-the-sustage-project">Supporting the Wellness at Work and Productivity of Ageing Employees in Industrial Environments: The sustAGE Project</a></p>
https://doi.org/10.5281/zenodo.4294256
oai:zenodo.org:4294256
eng
ERCIM
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.4294255
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
ERCIM News - Smart Things Everywere, 119, 40-41, (2019-10-08)
Supporting the Wellness at Work and Productivity of Ageing Employees in Industrial Environments: The sustAGE Project
info:eu-repo/semantics/technicalDocumentation
oai:zenodo.org:6610569
2022-06-03T11:06:56Z
user-sustage
user-eu
Gauri Deshpande
Björn W.Schuller
Anton Batliner
2021-08-30
<p>Abstract</p>
<p>The Coronavirus (COVID-19) pandemic impelled several research efforts, from collecting COVID-19 patients’ data to screening them for virus detection. Some COVID-19 symptoms are related to the functioning of the respiratory system that influences speech production; this suggests research on identifying markers of COVID-19 in speech and other human generated audio signals. In this article, we give an overview of research on human audio signals using ‘Artificial Intelligence’ techniques to screen, diagnose, monitor, and spread the awareness about COVID-19. This overview will be useful for developing automated systems that can help in the context of COVID-19, using non-obtrusive and easy to use bio-signals conveyed in human non-speech and speech audio productions.</p>
https://doi.org/10.5281/zenodo.6610569
oai:zenodo.org:6610569
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610568
info:eu-repo/semantics/restrictedAccess
Pattern Recognition, 122(February 2022), (2021-08-30)
COVID-19
Digital health
Audio processing
Computational paralinguistics
AI-Based human audio processing for COVID-19: A comprehensive overview
info:eu-repo/semantics/article
oai:zenodo.org:4294259
2020-12-09T10:55:09Z
user-sustage
Manolis Lourakis
Maria Pateraki
Ion-Anastasios Karolos
Christos Pikridas
Petros Patias
2020-08-12
<p><strong>Abstract.</strong> Without additional prior information, the pose of a camera estimated with computer vision techniques is expressed in a local coordinate frame attached to the camera’s initial location. Albeit sufficient in many cases, such an arbitrary representation is not convenient for employment in certain applications and has to be transformed to a coordinate system external to the camera before further use. Assuming a camera that is firmly mounted on a moving platform, this paper describes a method for continuously tracking the pose of that camera in a projected coordinate system. By combining exterior orientation from a known target with incremental pose changes inferred from accurate multi-GNSS positioning, the full 6 DoF pose of the camera is updated with low processing overhead and without requiring the continuous visual tracking of ground control points. Experimental results of applying the proposed method to a moving vehicle and a mobile port crane are reported, demonstrating its efficacy and potential.</p>
https://doi.org/10.5194/isprs-archives-XLIII-B2-2020-55-2020
oai:zenodo.org:4294259
eng
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLIII-B2-2020, 55–62, (2020-08-12)
ISPRS, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
pose
georeferencing
exterior orientation
absolute orientation
multi-GNSS
RTK
Pose estimation of a moving camera with low cost, Multi-GNSS devices
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:7043890
2022-09-06T14:26:24Z
openaire_data
user-sustage
Lourakis, Manolis
Pateraki, Maria
2022-09-02
<p>This dataset contains image sequences that feature a moving quay crane spreader in a port environment while unloading a container cargo vessel. A container crane spreader is a device that is installed on a crane and used to lift containers after attaching onto them.</p>
<p><br>
The sequences were acquired from a viewpoint similar to that of the crane operator using a camera installed next to the operator’s cabin at a height of approximately 20 meters above the quay. The camera thus moves with the crane, resulting in a non-stationary image background.</p>
<p>The dataset is organized into several RAR archives, one for each sequence. In addition to the undistorted image frames, it includes for every sequence a text file whose each line consists of the frame id for every image, the spreader’s bounding box and the spreader’s 6D pose (Rodrigues vector for the orientation, and the translation vector). The axis-aligned 2D bounding box is in the format <em>x0 y0 w h</em> where <em>(x0, y0)</em> is the top left corner and <em>w x h</em> its size, all in pixels. The spreader’s pose is defined with respect to the camera coordinate frame. Also included are the camera intrinsics matrix K for each sequence along with a common 3D mesh model for the spreader.</p>
<p>The spreader’s mesh model is supplied in PLY format. For a certain image frame, a model vertex M transforms to the camera coordinate system as R*M + t, R and t being the spreader’s pose (R is the equivalent rotation matrix). The homogeneous coordinates of that vertex’s projection on the image frame are K*(R*M + t).</p>
<p><br>
The dataset can support research on topics such as object localization, object detection, pose estimation, tracking, etc.<br>
If you use this dataset in your research work, you are kindly asked to cite the following paper in your publications:</p>
<p>M. Lourakis and M. Pateraki, "<em>Markerless Visual Tracking of a Container Crane Spreader,</em>" 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2021, pp. 2579-2586, doi: <a href="https://doi.org/10.1109/ICCVW54120.2021.00291">10.1109/ICCVW54120.2021.00291</a>.</p>
The sequences have been acquired by two different PtGrey color GigE cameras, FL3-GE-28S4C-C and BFLY-PGE-31S4C-C.
The frames of each sequence are in a separate RAR archive.
The spreader 3D model is in spreader40ft.ply.
https://doi.org/10.5281/zenodo.7043890
oai:zenodo.org:7043890
eng
Zenodo
https://zenodo.org/communities/sustage
https://doi.org/10.5281/zenodo.7043889
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
dataset
computer vision
tracking
pose estimation
object localization
object detection
Container spreader pose tracking dataset
info:eu-repo/semantics/other
oai:zenodo.org:6610509
2022-06-03T13:50:46Z
user-sustage
user-eu
Adria Mallol-Ragolta
Anastasia Semertzidou
Maria Pateraki
Björn Schuller
2022-03-22
<p>The advent of IoT devices in combination with Human Activity Recognition (HAR) technologies can contribute to battle with sedentariness by continuously monitoring the users' daily activities. With this information, autonomous systems could detect users' physical weaknesses and plan personalized training routines to improve them. This work investigates the multimodal fusion of smartwatch sensor data for HAR. Specifically, we exploit pedometer, heart rate, and accelerometer information to train unimodal and multimodal models for the task at hand. The models are trained end-to-end, and we compare the performance of dedicated Recurrent Neural Network-based (RNN) and Convolutional Neural Network-based (CNN) architectures to extract deep learnt representations from the input modalities. To fuse the embedded representations when training the multimodal models, we investigate a concatenation-based and an outer product-based approach. This work explores the harAGE dataset, a new dataset for HAR collected using a Garmin Vivoactive 3 device with more than 17 h of data. Our best models obtain an Unweighted Average Recall (UAR) of 95.6, 69.5, and 60.8% when tackling the task as a 2-class, 7-class, and 10-class classification problem, respectively. These performances are obtained using multimodal models that fuse the embedded representations extracted with dedicated CNN-based architectures from the pedometer, heart rate, and accelerometer modalities. The concatenation-based fusion scores the highest UAR in the 2-class classification problem, while the outer product-based fusion obtains the best performances in the 7-class and the 10-class classification problems.</p>
https://doi.org/10.5281/zenodo.6610509
oai:zenodo.org:6610509
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610508
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
Frontiers in Computer Science - Mobile and Ubiquitous Computing, (2022-03-22)
Outer Product-Based Fusion of Smartwatch Sensor Data for Human Activity Recognition
info:eu-repo/semantics/article
oai:zenodo.org:7092337
2022-09-20T02:26:33Z
user-sustage
user-eu
Manolis Lourakis
Maria Pateraki
2022-09-19
<p>Abstract: Workers in ports work with and in close proximity of heavy machinery. Quay cranes used for moving containers between ships and the dockside yard are one of the most accident-prone equipment types. For picking up containers, these cranes are equipped with spreaders, i.e. lifting devices which are lowered down on top of containers and lock on to them mechanically. We are concerned here with monitoring a moving quay crane spreader so as to make sure that safe clearance distances are maintained from the locations of dock workers in a port container cargo handling environment. The paper describes the application of computer vision techniques to develop a model-based, monocular spreader tracker. By tracking in three dimensions the position and orientation of the spreader during loading and unloading operations, a threat volume enclosing it can be defined. Constantly monitoring the distance of dock workers from this threat volume can improve the operator’s situational awareness and increase safety in the work environment. Quantitative experimental evaluation is also reported.</p>
https://doi.org/10.54941/ahfe1002146
oai:zenodo.org:7092337
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
AHFE 2022, International Conference on Applied Human Factors and Ergonomics, New York, USA, 24-28 july 2022
Occupational Safety
Quay crane
container
computer vision
tracking
Computer vision for increasing safety in container handling operations
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6610264
2022-06-03T09:14:54Z
user-sustage
user-eu
Philipp Klumpp
Tomas Arias-Vergara
Juan Camilo Vasquez-Correa
Paula Andrea
Perez-Toro
Juan Rafael Orozco-Arroyave
Anton Batliner
Elmar Noth
2021-11-18
<p>Abstract</p>
<p>As one of the most prevalent neurodegenerative disorders, Parkinson’s disease (PD) has a significant impact on the fine motor skills of patients. The complex <a href="https://www.sciencedirect.com/topics/engineering/interplay">interplay</a> of different <a href="https://www.sciencedirect.com/topics/engineering/articulator">articulators</a> during speech production and realization of required muscle tension become increasingly difficult, thus leading to a dysarthric speech. Characteristic patterns such as vowel instability, slurred pronunciation and slow speech can often be observed in the affected individuals and were analyzed in previous studies to determine the presence and progression of PD. In this work, we used a phonetic <a href="https://www.sciencedirect.com/topics/computer-science/recognizers">recognizer</a> trained exclusively on healthy speech data to investigate how PD affected the phonetic footprint of patients. We rediscovered numerous patterns that had been described in previous contributions although our system had never seen any pathological speech previously. Furthermore, we could show that intermediate activations from the <a href="https://www.sciencedirect.com/topics/social-sciences/neural-network">neural network</a> could serve as feature vectors encoding information related to the disease state of individuals. We were also able to directly correlate the expert-rated intelligibility of a speaker with the mean confidence of phonetic predictions. Our results support the assumption that pathological data is not necessarily required to train systems that are capable of analyzing PD speech.</p>
https://doi.org/10.5281/zenodo.6610264
oai:zenodo.org:6610264
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610263
info:eu-repo/semantics/restrictedAccess
Computer Speech & Language, 72, (2021-11-18)
Parkinson's disease
Phonetic analysis
Phoneme recognition
Pathological speech
The phonetic footprint of Parkinson's disease
info:eu-repo/semantics/article
oai:zenodo.org:5806100
2022-06-03T10:21:34Z
user-sustage
user-eu
G Athanassiou
P Gajewski
A Ascolese
S Ballesteros
M Maniadakis
M Pateraki
A Prieto
I Varlamis
R Monferino
2021-09-02
<p>Purpose Cognitive ageing may impair the ability of older employees to remain in the workforce. The interdisciplinary EU-project sustAGE draws upon digital technology trends to support older workers by providing guidance for health-promoting activities in and outside the working context. One of sustAGE’s objectives is the enhancement of age-affected cognitive abilities through targeted training interventions. A combination of personalized recommendations and corresponding cognitive trainings in form of digital serious games serves this purpose. Three interlinked steps are necessary for implementing the sustAGE approach: base-line assessment of cognitive abilities of potential groups of users; derivation and implementation of user-specific recommendations; and development and evaluation of training. The first step within sustAGE development included an empirical assessment of cognitive abilities from a sample of older employees in two of the most important sectors of EU industry: automotive industry (AI) and maritime logistics (ML).</p>
<p>Methods A psychometric test battery measuring attentional, memory and executive functions was administered to a sample of 60 older employees (M = 53.4; SD = 5.1) from the two industries.</p>
<p>Results Comparisons between occupational groups revealed significant effects regarding cognitive performance (AI > HL; sustained attention, p < .0001; processing speed, p < .005; task switching, p < .0001).</p>
<p>Conclusions Group differences in cognitive performance may relate to specific job characteristics of the two occupational domains and provide evidence for differentiated, group-specific needs and respective interventions. Current results as well as additional sample analysis will serve as basis for the development of targeted recommendations and cognitive trainings based on the inferred group profiles.</p>
https://doi.org/10.1055/s-0041-1731984
oai:zenodo.org:5806100
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/restrictedAccess
Gesundheitswesen, 83(08/09), 249, (2021-09-02)
Sustainable work through technology-assisted enhancement of cognitive abilities of older employees: the sustAGE approach
info:eu-repo/semantics/article
oai:zenodo.org:4294212
2020-11-28T12:27:16Z
user-sustage
user-eu
Adria Mallol-Ragolta
Nicholas Cummins
Björn Schuller
2020-10-29
<p>One of the keys for supervised learning techniques to succeed resides in the access to vast amounts of labelled training data. The process of data collection, however, is expensive, time- consuming, and application dependent. In the current digital era, data can be collected continuously. This continuity renders data annotation into an endless task, which potentially, in problems such as emotion recognition, requires annotators with different cultural backgrounds. Herein, we study the impact of utilising data from different cultures in a semi-supervised learning ap- proach to label training material for the automatic recognition of arousal and valence. Specifically, we compare the performance of culture-specific affect recognition models trained with man- ual or cross-cultural automatic annotations. The experiments performed in this work use the dataset released for the Cross- cultural Emotion Sub-challenge of the Audio/Visual Emotion Challenge (AVEC) 2019. The results obtained convey that the cultures used for training impact on the system performance. Furthermore, in most of the scenarios assessed, affect recogni- tion models trained with hybrid solutions, combining manual and automatic annotations, surpass the baseline model, which was exclusively trained with manual annotations.</p>
https://doi.org/10.21437/Interspeech.2020-2641
oai:zenodo.org:4294212
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
INTERSPEECH 2020, Shanghai, China, October 25–29, 2020
continuous affect recognition, cross-cultural anal- ysis, audiovisual processing, semi-supervised learning
An Investigation of Cross-Cultural Semi-Supervised Learning for Continuous Affect Recognition
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4294179
2020-12-08T14:21:34Z
user-sustage
user-eu
Lukas Stappen
Vincent Karas
Nicholas Cummins
Fabien Ringeval
Klaus Scherer
Björn Schuller
2019-11-18
<p><strong>Abstract</strong></p>
<p>Multimodal data sources offer the possibility to capture and model interactions between modalities, leading to an improved understanding of underlying relationships. In this regard, the work presented in this paper explores the relationship between facial muscle movements and speech signals. Specifically, we explore the efficacy of different sequence-to-sequence neural network architectures for the task of predicting Facial Action Coding System Action Units (AUs) from one of two acoustic feature representations extracted from speech signals, namely the extended Geneva Minimalistic Acoustic Parameter Set (eGeMAPs) or the Interspeech Computational Paralinguistics Challenge features set (ComParE). Furthermore, these architectures were enhanced by two different attention mechanisms (intra- and inter-attention) and various state-of-the-art network settings to improve prediction performance. Results indicate that a sequence-to-sequence model with inter-attention can achieve on average an Unweighted Average Recall (UAR) of 65.9 % for AU onset, 67.8 % for AU apex (both eGeMAPs), 79.7 % for AU offset and 65.3 % for AU occurrence (both ComParE) detection over all AUs.</p>
2019 IEEE 21st International Workshop on Multimedia Signal Processing (MMSP)
DOI: 10.1109/MMSP46350.2019
Funding : BMW Group Research
https://doi.org/10.1109/MMSP.2019.8901779
oai:zenodo.org:4294179
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/restrictedAccess
MMSP, IEEE 21st International Workshop on Multimedia Signal Processing, Kuala Lumpur, Malaysia, 27-29 Sept. 2019
attention networks
facial action units
sequence to sequence
paralingustics
From Speech to Facial Activity: Towards Cross-modal Sequence-to-Sequence Attention Networks
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:6610291
2022-06-03T09:22:26Z
user-sustage
user-eu
Anton Batliner
Simone Hantke
Bjoern W. Schuller
2022-06-03
<p>Abstract:</p>
<p>With the advent of ‘heavy Artificial Intelligence’ - big data, deep learning, and ubiquitous use of the internet, ethical considerations are widely dealt with in public discussions and governmental bodies. Within Computational Paralinguistics with its manifold topics and possible applications (modelling of long-term, medium-term, and short-term traits and states such as personality, emotion, or speech pathology), we have not yet seen that many contributions. In this article, we try to set the scene by (1) giving a short overview of ethics and privacy, (2) describing the field of Computational Paralinguistics, its history and exemplary use cases, as well as (de-)anonymisation and peculiarities of speech and text data, and (3) proposing rules for good practice in the field, such as choosing the right performance measure, and accounting for representativity and interpretability.</p>
https://doi.org/10.5281/zenodo.6610291
oai:zenodo.org:6610291
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.6610290
info:eu-repo/semantics/restrictedAccess
IEEE Transactions on Affective Computing, 1949-3045, (2022-06-03)
Ethics
Privacy
Data privacy
Medical services
Guidelines
Affective computing
Big Data
Ethics and Good Practice in Computational Paralinguistics
info:eu-repo/semantics/article
oai:zenodo.org:4106434
2020-11-28T12:27:15Z
user-sustage
user-eu
A. Mallol-Ragolta
Z. Zhao
L. Stappen
N. Cummins
B. Schuller
2019-09-19
<p>The high prevalence of depression in society has given rise to a need for new digital tools that can aid its early detection. Among other effects, depression impacts the use of language. Seeking to exploit this, this work focuses on the detection of depressed and non-depressed individuals through the analysis of linguistic information extracted from transcripts of clinical interviews with a virtual agent. Specifically, we investigated the advantages of employing hierarchical attention-based networks for this task. Using Global Vectors (GloVe) pretrained word embedding models to extract low-level representations of the words, we compared hierarchical local-global attention networks and hierarchical contextual attention networks. We performed our experiments on the Distress Analysis Interview Corpus - Wizard of Oz (DAIC-WoZ) dataset, which contains audio, visual, and linguistic information acquired from participants during a clinical session. Our results using the DAIC-WoZ test set indicate that hierarchical contextual attention networks are the most suitable configuration to detect depression from transcripts. The configuration achieves an Unweighted Average Recall (UAR) of .66 using the test set, surpassing our baseline, a Recurrent Neural Network that does not use attention.</p>
Funding by EU- sustAGE (826506), EU-RADAR-CNS (115902), Key Program of the Natural Science Foundation of Tianjin, CHINA (18JCZDJC36300) and BMW Group Research
Pages 221-225
https://www.isca-speech.org/archive/Interspeech_2019/index.html
https://doi.org/10.21437/Interspeech.2019-2036
oai:zenodo.org:4106434
eng
Zenodo
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
INTERSPEECH 2019, 20th Annual Conference of the International Speech Communication Association, GRaz, Austria, 15-19 September 2019
A Hierarchical Attention Network-Based Approach for Depression Detection from Transcribed Clinical Interviews. Proc. Interspeech 2019
info:eu-repo/semantics/conferencePaper
oai:zenodo.org:4294279
2020-12-08T14:38:13Z
user-sustage
Mónica Muiños
Soledad Ballesteros
2020-04-14
<p>T’ai Chi has become increasingly popular as an activity, involving gentle movements and relaxation. Nowadays it is widely accepted that martial arts improve physical, cognitive, and psychological health, contributing to general well-being. This chapter examines recent findings related to the effects of T’ai Chi on the cognition and brain functions of older adults. It provides a discussion on the eye-hand coordination, visuospatial abilities, and visual asymmetries. The brain adapts to changes demanded by the environment, permitting learning throughout life despite cognitive slowing, even at a very old age. The chapter summarizes the main results of recent neuroimaging studies investigating the effects of T’ai Chi on the brain. In a research study, the T’ai Chi and endurance exercise groups outperformed the older sedentary adults; the latter had longer reaction times on the behavioural task and lower P3 amplitude.</p>
https://doi.org/10.4324/9781315187228
oai:zenodo.org:4294279
eng
Routledge
https://zenodo.org/communities/sustage
info:eu-repo/semantics/restrictedAccess
T´ai Chi to improving brain and cognition
info:eu-repo/semantics/bookPart
oai:zenodo.org:5801456
2021-12-27T13:48:41Z
user-sustage
María Pilar Jiménez
Jennifer A. Rieker
José Manuel Reales
Soledad Ballesteros
2021-05-14
<p>Abstract:</p>
<p>The sudden outbreak of the COVID-19 pandemic has profoundly altered the daily lives of the population with dramatic effects caused not only by the health risks of the coronavirus, but also by its psychological and social impact in large sectors of the worldwide population. The present study adapted the COVID-19 Peritraumatic Distress Index (CPDI) to the Spanish population, and 1094 Spanish adults (mean age 52.55 years, 241 males) completed the Spanish version in a cross-sectional online survey. To analyze the factorial structure and reliability of the CPDI, we performed an exploratory factor analysis (EFA) followed by a confirmatory factor analysis (CFA) on the Spanish sample. The effects of gender and age on the degree of distress were analyzed using the factorial scores of the CPDI as the dependent variables. Results showed that, after rotation, the first factor (Stresssymptoms) accounted for 35% of the total variance and the second factor (COVID-19 information) for 15%. Around 25% (n = 279) of the participants experienced mild to moderate distress symptoms, 16% (n = 179) severe distress, and about 58% (n = 636) showed no distress symptoms. Women experienced more distress than men (<em>p</em><0.01), and distress decreased with age (<em>p</em><0.01). We conclude that the CPDI seems a promising screening tool for the rapid detection of potential peritraumatic stress caused by the COVID-19 pandemic.</p>
<p> </p>
<p> </p>
<p> </p>
<p> </p>
https://doi.org/10.3390/ijerph18105253
oai:zenodo.org:5801456
Zenodo
https://zenodo.org/communities/sustage
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
International Journal of Environmental Research and Public Health, 8(10), (2021-05-14)
COVID-19
Peritraumatic Distress Index
cluster analysis
exploratory factor analysis (EFA)
onfirmatory factor analysis (CFA)
age differences
psychological distress
psychological impact
COVID-19 Peritraumatic Distress as a Function of Age and Gender in a Spanish Sample
info:eu-repo/semantics/article
oai:zenodo.org:3609982
2020-11-28T00:26:56Z
user-sustage
user-eu
Maria Pateraki
Konstantinos Fysarakis
Vangelis Sakkalis
Georgios Spanoudakis
Iraklis Varlamis
Michail Maniadakis
Manolis Lourakis
Sotiris Ioannidis
Nicholas Cummins
Björn Schuller
Evangelos Loutsetis
Dimitrios Koutsouris
2019-09-13
<p>The use of health and well-being monitoring technologies has been steadily increasing and such systems can now be found in smart homes, age-friendly workplaces, public spaces, and elsewhere. These monitoring technologies employ a wide variety of off-the-shelf smart sensors and medical devices to support functional, physiological, and behavioral monitoring and to address social interaction aspects of daily life. These systems focus either on specific health-related conditions or on supporting the more general aims of comfort, well-being, and quality of life. However, there remain several technological (interoperability, expandability, etc.) and societal (cost, privacy, etc.) challenges to be addressed before smart biosensor systems are widely adopted.</p>
<p>Motivated by the above, this chapter highlights the challenges and opportunities surrounding the application of smart biosensors in healthcare and presents three state of the art solutions for leveraging smart sensors in this context. The first concerns a <em>smart living solution platform</em> that integrates heterogeneous sensors and assistive medical and mobile devices, enabling continuous data collection from the everyday lives of the elderly and data analytics that support personalized interventions. The second presents an <em>Internet of Things (IoT) ecosystem</em> comprising sensors and smart wearables to improve occupational safety and workforce productivity through personalized recommendations. The last case is an <em>intelligent noninvasive biosignal recording system</em> that detects potentially hazardous pathological conditions of infants during sleep.</p>
https://doi.org/10.1016/B978-0-12-815369-7.00002-1
oai:zenodo.org:3609982
eng
Elsevier
https://zenodo.org/communities/sustage
https://zenodo.org/communities/eu
info:eu-repo/semantics/restrictedAccess
Internet of Things, attribute-based encryption, hazardous situations detection, temporal abstraction, elderly monitoring, infant monitoring, artificial intelligence
Biosensors and Internet of Things in smart healthcare applications: challenges and opportunities
info:eu-repo/semantics/bookPart