Action recognition using invariant features under unexampled viewing conditions

Litian Sun, Kiyoharu Aizawa

Abstract

A great challenge in real-world applications of action recognition is the lack of sufficient label information because of variance in the recording viewpoint and differences between individuals. A system that can adapt itself according to these variances is required for practical use. We present a generic method for extracting view-invariant features from skeleton joints. These view-invariant features are further refined using a stacked, compact autoencoder. To model the challenge of real-world applications, two unexampled test settings (NewView and NewPerson) are used to evaluate the proposed method. Experimental results with these test settings demonstrate the effectiveness of our method.
BibTeX
@inproceedings{sun2013action,
  author = {Litian Sun and Kiyoharu Aizawa},
  title = {{Action recognition using invariant features under unexampled viewing conditions}},
  booktitle = {ACM Multimedia},
  year = {2013},
  doi = {10.1145/2502081.2508126},
}

Loading related content...

Figure