Sciweavers

Share
CVPR
2006
IEEE

Unsupervised Discovery of Action Classes

12 years 10 months ago
Unsupervised Discovery of Action Classes
In this paper we consider the problem of describing the action being performed by human figures in still images. We will attack this problem using an unsupervised learning approach, attempting to discover the set of action classes present in a large collection of training images. These action classes will then be used to label test images. Our approach uses the coarse shape of the human figures to match pairs of images. The distance between a pair of images is computed using a linear programming relaxation technique. This is a computationally expensive process, and we employ a fast pruning method to enable its use on a large collection of images. Spectral clustering is then performed using the resulting distances. We present clustering and image labeling results on a variety of datasets.
Greg Mori, Hao Jiang, Mark S. Drew, Yang Wang 0003
Added 12 Oct 2009
Updated 12 Oct 2009
Type Conference
Year 2006
Where CVPR
Authors Greg Mori, Hao Jiang, Mark S. Drew, Yang Wang 0003, Ze-Nian Li
Comments (0)
books