2010. 6. 19. 17:06 Paper Reading/CVPR
SPEC Hashing: Similarity Preserving algorithm for Entropy-based Coding,
Ruei-Sung Lin, David Ross, Jay Yagnik. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2010.
http://www.cs.toronto.edu/~dross/LinRossYagnik_CVPR2010.pdf
Abstract:
Searching approximate nearest neighbors in large scale high dimensional data set has been a challenging problem . This paper presents a novel and fast algorithm for learning binary hash functions for fast nearest neighbor retrieval. The nearest neighbors are defined according to the semantic similarity between the objects. Our method uses the information of these semantic similarities and learns a hash function with binary code such that only objects with high similarity have small Hamming distance. The hash function is incrementally trained one bit at a time, and as bits are added to the hash code Hamming distances between dissimilar objects increase. We further link our method to the idea of maximizing conditional entropy among pair of bits and derive an extremely efficient linear time hash learning algorithm. Experiments on similar image retrieval and celebrity face recognition show that our method produces apparent improvement in performance over some state-of-the-art methods.
http://www.cs
Abstract:
Se
'Paper Reading > CVPR' 카테고리의 다른 글
Microsoft Research Street Slide View (0) | 2010.07.30 |
---|---|
Interest Seam Image (0) | 2010.07.28 |
Faster-than-SIFT Object Detection (0) | 2010.06.18 |
cvpr2010 papers top #10 (0) | 2010.05.30 |
INRIA (0) | 2010.05.16 |