Markatopoulou, Foteini
Moumtzidou, Anastasia
Galanopoulos, Damianos
Avgerinakis, Konstantinos
Andreadis, Stelios
Gialampoukidis, Ilias
Tachos, Stavros
Vrochidis, Stefanos
Mezaris, Vasileios
Kompatsiaris, Ioannis
Patras, Ioannis
2018-02-23
<p>This paper provides an overview of the runs submitted to TRECVID 2017 by ITI-CERTH. ITI-CERTH participated in the Ad-hoc Video Search (AVS), Multimedia Event Detection (MED), Instance Search (INS) and Surveillance Event Detection (SED) tasks. Our AVS task participation is based on a method that combines the linguistic analysis of the query with concept-based and semantic-embedding representations of video fragments. Regarding the MED task, this year we participate on Pre-Specied and Ah-Hoc sub-tasks exploiting both motion-based as well as DCNN-based features. The INS task is performed by employing VERGE, which is an interactive retrieval application that integrates retrieval functionalities that consider mainly visual information. For the SED task, we deploy a novel activity detection algorithm that is based on human detection in video frames, goal descriptors, dense trajectories, Fisher vectors and a discriminative action segmentation scheme.</p>
https://doi.org/10.5281/zenodo.1183440
oai:zenodo.org:1183440
Zenodo
https://zenodo.org/communities/invid-h2020
https://zenodo.org/communities/emma-h2020
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.1183439
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
TRECVID 2017 Workshop, Gaithersburg, MD, USA, November 2017
ITI-CERTH participation in TRECVID 2017
info:eu-repo/semantics/conferencePaper