Patch-based image classification using image epitomes

Training Collage

The goal of this project is to design a classifier capable of learning to recognize whether an image belongs to a certain class or not.  Example classes would be "human face", "beach", "tiger", etc.  The classifier would be given a series of training images and told which are positive and negative examples for the classification.  From these training examples, the classifier would extract common features that could be used to decide whether a new image belongs to the class or not.  

This project attempts to acheive this by finding patches that are common to all or most of the positive example images, but absent from all or most of the negative example images.  These patches could be called "discriminative patches" because they discriminate between positive and negative examples.  

In order to find these discriminative patches, this project uses a probabilistic image representation known as an "image epitome".  This representation condenses an image into a smaller collection of patches that can be quilted  to reconstructed the original image.  

The basic method employed by this project is to
Presentation (ppt)

Final report (pdf, html)

Experimental results

References
Image Epitomes
N. Jojic, B. J. Frey, and A. Kannan, Epitomic analysis of appearance and shape, Proc. 9th Int. Conf. Computer Vision, 2003. (webpage)

Video Epitomes
V. Cheung, B. J. Frey, and N. Jojic, Video epitomes, Proc. IEEE Conf. Computer Vision and Pattern Recognition, 2005. (webpage)

Object Recognition
R. Fergus, P. Perona, A. Zisserman, Object Class Recognition by Unsupervised Scale-Invariant Learning, Proc. of the IEEE Conf on Computer Vision and Pattern Recognition, 2003.