Poster
Heavy-tailed Distances for Gradient Based Image Descriptors
Yangqing Jia · Trevor Darrell

Wed Dec 14th 05:45 -- 11:59 PM @ None #None
in Poster Session »

Many applications in computer vision measure the similarity between images or image patches based on some statistics such as oriented gradients. These are often modeled implicitly or explicitly with a Gaussian noise assumption, leading to the use of the Euclidean distance when comparing image descriptors. In this paper, we show that the statistics of gradient based image descriptors often follow a heavy-tailed distribution, which undermines any principled motivation for the use of Euclidean distances. We advocate for the use of a distance measure based on the likelihood ratio test with appropriate probabilistic models that fit the empirical data distribution. We instantiate this similarity measure with the Gamma-compound-Laplace distribution, and show significant improvement over existing distance measures in the application of SIFT feature matching, at relatively low computational cost.

Author Information

Yangqing Jia (Alibaba Group)
Trevor Darrell (UC Berkeley)

More from the Same Authors