Conference paper Open Access

Short-term Recognition of Human Activities using Convolutional Neural Networks

M.Papakostas; T. Giannakopoulos; F. Makedon; V. Karkaletsis

Citation Style Language JSON Export

  "publisher": "Zenodo", 
  "DOI": "10.5281/zenodo.376482", 
  "title": "Short-term Recognition of Human Activities using Convolutional Neural Networks", 
  "issued": {
    "date-parts": [
  "abstract": "<p>This paper proposes a deep learning classification method for frame-wise recognition of human activities, using raw color (RGB) information. In particular, we\u00a0present a Convolutional Neural Network (CNN) classification\u00a0approach for recognising three basic motion activity classes, that cover the vast majority of human activities in the context\u00a0of a home monitoring environment, namely: sitting, walking\u00a0and standing up. A real-world fully annotated dataset has\u00a0been compiled, in the context of an assisted living home\u00a0environment. Through extensive experimentation we have\u00a0highlighted the benefits of deep learning architectures against\u00a0traditional shallow classifiers functioning on hand-crafted\u00a0features, on the task of activity recognition. Our approach\u00a0proves the robustness and the quality of CNN classifiers\u00a0that lies on learning highly invariant features. Our ultimate\u00a0goal is to tackle the challenging task of activity recognition\u00a0in environments that are characterized with high levels of\u00a0inherent noise.</p>", 
  "author": [
      "family": "M.Papakostas"
      "family": "T. Giannakopoulos"
      "family": "F. Makedon"
      "family": "V. Karkaletsis"
  "id": "376482", 
  "type": "paper-conference", 
  "event": "2016 International Conference on Signal-Image Technology & Internet Based Systems, IEEE, 2016 (SITIS)"
All versions This version
Views 133133
Downloads 145145
Data volume 230.5 MB230.5 MB
Unique views 117117
Unique downloads 140140


Cite as