Colour in Context
Research group
Computer Vision Center

Coloring Action Recognition in Still Images

Fahad Shahbaz Khan, Muhammad Anwer Rao, Joost van de Weijer, Andrew D. Bagdanov, Antonio Lopez, Michael Felsberg
International Journal of Computer Vision (IJCV), Volume 105, Number 3, page 205--221 - dec 2013
IF: 3.533. area: COMPUTER SCIENCE, ARTIFICIAL I. Quartile: 1.
Download the publication : IJCV2013.pdf [7.2Mo]  
In this article we investigate the problem of human action recognition in static images. By ac- tion recognition we intend a class of problems which includes both action classi cation and action detection (i.e. simultaneous localization and classi cation). Bag- of-words image representations yield promising results for action classi cation, and deformable part models perform very well object detection. The representations for action recognition typically use only shape cues and ignore color information. Inspired by the recent success of color in image classi cation and object detection, we investigate the potential of color for action classi cation and detection in static images. We perform a comprehensive evaluation of color de- scriptors and fusion approaches for action recognition. Experiments were conducted on the three datasets most used for benchmarking action recognition in still im- ages: Willow, PASCAL VOC 2010 and Stanford-40. Our experiments demonstrate that incorporating color information considerably improves recognition perfor- mance, and that a descriptor based on color names outperforms pure color descriptors. Our experiments demonstrate that late fusion of color and shape infor- mation outperforms other approaches on action recog- nition. Finally, we show that the di erent color-shape fusion approaches result in complementary information and combining them yields state-of-the-art performance for action classi cation.

Images and movies

 

BibTex references

@Article\{SRV2013,
  author       = "Fahad Shahbaz Khan and Muhammad Anwer Rao and Joost van de Weijer and Andrew D. Bagdanov and Antonio Lopez and Michael Felsberg",
  title        = "Coloring Action Recognition in Still Images",
  journal      = "International Journal of Computer Vision (IJCV)",
  number       = "3",
  volume       = "105",
  pages        = "205--221",
  month        = "dec",
  year         = "2013",
  abstract     = "In this article we investigate the problem
of human action recognition in static images. By ac-
tion recognition we intend a class of problems which
includes both action classication and action detection
(i.e. simultaneous localization and classication). Bag-
of-words image representations yield promising results
for action classication, and deformable part models
perform very well object detection. The representations
for action recognition typically use only shape cues and
ignore color information. Inspired by the recent success
of color in image classication and object detection, we
investigate the potential of color for action classication
and detection in static images.
We perform a comprehensive evaluation of color de-
scriptors and fusion approaches for action recognition.
Experiments were conducted on the three datasets most
used for benchmarking action recognition in still im-
ages: Willow, PASCAL VOC 2010 and Stanford-40.
Our experiments demonstrate that incorporating color
information considerably improves recognition perfor-
mance, and that a descriptor based on color names
outperforms pure color descriptors. Our experiments
demonstrate that late fusion of color and shape infor-
mation outperforms other approaches on action recog-
nition. Finally, we show that the dierent color-shape
fusion approaches result in complementary information
and combining them yields state-of-the-art performance
for action classication.",
  ifactor      = "3.533",
  quartile     = "1",
  area         = "COMPUTER SCIENCE, ARTIFICIAL I",
  url          = "http://cat.uab.es/Public/Publications/2013/SRV2013"
}

Other publications in the database

  2008 Colour in context Group | Computer Vision Center. All rights reserved | Contact webmaster |  Last updated: Monday 11 May 2009     eXTReMe Tracker