%0 Conference Proceedings %T Packing bag-of-features %+ Learning and recognition in vision (LEAR) %+ Multimedia content-based indexing (TEXMEX) %A Jégou, Hervé %A Douze, Matthijs %A Schmid, Cordelia %Z RAFFUT, QUAERO %< avec comité de lecture %B ICCV 2009 - 12th International Conference on Computer Vision %C Kyoto, Japan %I IEEE Computer society %P 2357-2364 %8 2009-09-29 %D 2009 %R 10.1109/ICCV.2009.5459419 %K data compression %K image coding %K indexing %Z Computer Science [cs]/Information Retrieval [cs.IR] %Z Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] %Z Computer Science [cs]/Databases [cs.DB]Conference papers %X One of the main limitations of image search based on bag-of-features is the memory usage per image, limiting to a few million the size of the dataset that can be handled on a single machine in a reasonable response time. In this paper, we first show that these limitations can be somewhat reduced by using index compression. Then, we propose an image representation obtained by projecting bag-of-features histograms onto a set of predefined sparse projection functions, producing several image descriptors. Coupled with a proper indexing structure, an image is represented by a few hundred bytes. A distance expectation criterion is then used to rank the images. Our method is at least one order of magnitude faster than standard bag-of-features while providing excellent search quality. %G English %2 https://inria.hal.science/inria-00394213/document %2 https://inria.hal.science/inria-00394213/file/jegou_packingbof.pdf %L inria-00394213 %U https://inria.hal.science/inria-00394213 %~ EC-PARIS %~ UNIV-RENNES1 %~ UGA %~ CNRS %~ INRIA %~ UNIV-GRENOBLE1 %~ UNIV-PMF_GRENOBLE %~ INSA-RENNES %~ INPG %~ INRIA-RHA %~ INRIA-RENNES %~ IRISA %~ IRISA_SET %~ INRIA_TEST %~ LJK %~ LJK_GI %~ LJK_GI_LEAR %~ TESTALAIN1 %~ IRISA-D6 %~ UGA-TEST-QUATER %~ INRIA2 %~ UR1-HAL %~ UR1-MATH-STIC %~ UR1-UFR-ISTIC %~ TEST-UNIV-RENNES %~ TEST-UR-CSS %~ UNIV-RENNES %~ INRIA-RENGRE %~ INRIA-300009 %~ INSA-GROUPE %~ ANR %~ UR1-MATH-NUM