Packing bag-of-features

One of the main limitations of image search based on bag-of-features is the memory usage per image. Only a few million images can be handled on a single machine in rea- sonable response time. In this paper, we first evaluate how the memory usage is reduced by using lossless index com- pression. We then propose an approximate representation of bag-of-features obtained by projecting the corresponding histogram onto a set of pre-defined sparse projection func- tions, 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 mag- nitude faster than standard bag-of-features while providing excellent search quality.

• Type: Conference
• Year: 2009
• Where: ICCV
• Authors: Herve Jegou, Matthijs Douze, Cordelia Schmid
• Keywords: Computer Vision, Feature Extraction, ICCV 2009