High-confidence near-duplicate image detection

Wei Dong, Zhe Wang, Moses Charikar, Kai Li
2012 Proceedings of the 2nd ACM International Conference on Multimedia Retrieval - ICMR '12  
In this paper, we propose two techniques for near-duplicate image detection at high confidence and large scale. First, we show that entropy-based filtering eliminates ambiguous SIFT features that cause most of the false positives, and enables claiming nearduplicity with a single match of the retained high-quality features. Second, we show that graph cut can be used for query expansion with a duplicity graph computed offline to substantially improve search quality. Evaluation with web images
more » ... that when combined with sketch embedding [6], our methods achieve false positive rate orders of magnitude lower than the standard visual word approach. We demonstrate the proposed techniques with a largescale image search engine which, using indexing data structure offline computed with a Hadoop cluster, is capable of serving more than 50 million web images with a single commodity server.
doi:10.1145/2324796.2324798 dblp:conf/mir/DongWCL12 fatcat:ox2eabongveypa2cglvy7l3jxm