Discriminative Decorrelation for Clustering and Classification^*

Bharath Hariharan¹, Jitendra Malik¹, and Deva Ramanan²

¹Univerisity of California at Berkeley, Berkeley, CA, USA
bharath2@cs.berkeley.edu
malik@cs.berkeley.edu

²University of California at Irvine, Irvine, CA, USA
dramanan@ics.uci.edu

Abstract. Object detection has over the past few years converged on using linear SVMs over HOG features. Training linear SVMs however is quite expensive, and can become intractable as the number of categories increase. In this work we revisit a much older technique, viz. Linear Discriminant Analysis, and show that LDA models can be trained almost trivially, and with little or no loss in performance. The covariance matrices we estimate capture properties of natural images. Whitening HOG features with these covariances thus removes naturally occuring correlations between the HOG features. We show that these whitened features (which we call WHO) are considerably better than the original HOG features for computing similarities, and prove their usefulness in clustering. Finally, we use our findings to produce an object detection system that is competitive on PASCAL VOC 2007 while being considerably easier to train and test.

^*This work was funded by ONR-MURI Grant N00014-10-1-0933 and NSF Grant 0954083.

LNCS 7575, p. 459 ff.

Full article in PDF | BibTeX

lncs@springer.com
© Springer-Verlag Berlin Heidelberg 2012