Asymmetric Hamming Embedding

Mihir Jain 1 Hervé Jégou 1 Patrick Gros 1
1 TEXMEX - Multimedia content-based indexing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : This paper proposes an asymmetric Hamming Embedding scheme for large scale image search based on local descriptors. The comparison of two descriptors relies on an vector-to-binary code comparison, which limits the quantization error associated with the query compared with the original Hamming Embedding method. The approach is used in combination with an inverted file structure that offers high efficiency, comparable to that of a regular bag-of-features retrieval systems. The comparison is performed on two popular datasets. Our method consistently improves the search quality over the symmetric version. The trade-off between memory usage and precision is evaluated, showing that the method is especially useful for short binary signatures.
Document type :
Conference papers
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download


https://hal.inria.fr/inria-00607278
Contributor : Hervé Jégou <>
Submitted on : Friday, July 8, 2011 - 2:05:04 PM
Last modification on : Friday, November 16, 2018 - 1:21:44 AM

Files

hal.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00607278, version 2

Citation

Mihir Jain, Hervé Jégou, Patrick Gros. Asymmetric Hamming Embedding. ACM Multimedia, Nov 2011, Scottsdale, United States. ⟨inria-00607278v2⟩

Share

Metrics

Record views

744

Files downloads

893