Action Recognition in Still Images
We investigate the problem of action recognition in static images, for which problem bag-of-words image representations obtain promising results. These representations typically use only shape cues and ignore color information. Inspired by the recent success of color in image classification and object recognition, we investigate the potential of color for action recognition in static images.
We perform a comprehensive evaluation of state-of-the-art color descriptors
and fusion approaches for action recognition. Experiments were conducted
on the three principal datasets for benchmarking of action recognition
in still images: Willow, PASCAL VOC 2010 and Stanford-40. Our experiments
demonstrate that incorporating color information considerably improves
recognition performance. A descriptor based on color names is shown to
outperform other pure color descriptors. Differently than in image classification,
where early fusion yields the best results, our experiments demonstrate
that late fusion of color and shape information outperforms other approaches
on action recognition. Finally, we show that the different color-shape
fusion approaches result in complementary information and combining them
yields state-of-the-art performance.
F Coloring Action Recognition in Still Images , International Journal in Computer Vison (IJCV), 105(3):205:221, 2013.
-
Fahad Shahbaz Khan, Rao Muhammad Anwer, Joost van de Weijer, Andrew Bagdanov, Maria Vanrell, Antonio M. Lopez, Color Attributes for Object Detection, Proc. (CVPR)( webpage+code)
-
Fahad Shahbaz Khan, Joost van de Weijer, Andrew Bagdanov, Maria Vanrell, Portmanteau Vocabularies for Multi-Cue Image Representation, Proc. (NIPS)( webpage+code)
-
Modulating Shape Features by Color Attention for Object Recognition , International Journal of Computer Vision (IJCV), vol 98(1), 49-64, 2012.( webpage+code)
The article is partially based on earlier publications: