Searching in one billion vectors: re-rank with source coding

Details Searching in one billion vectors: re-rank with source coding Searching in one billion vectors: re-rank with source coding

2011-02-18

arxiv.org/abs/1102.3828v1

Computer Science

Information Retrieval

International Conference on Acoustics, Speech and Signal Processing, Prague : Czech Republic (2011)

Scientific paper

Recent indexing techniques inspired by source coding have been shown successful to index billions of high-dimensional vectors in memory. In this paper, we propose an approach that re-ranks the neighbor hypotheses obtained by these compressed-domain indexing methods. In contrast to the usual post-verification scheme, which performs exact distance calculation on the short-list of hypotheses, the estimated distances are refined based on short quantization codes, to avoid reading the full vectors from disk. We have released a new public dataset of one billion 128-dimensional vectors and proposed an experimental setup to evaluate high dimensional indexing algorithms on a realistic scale. Experiments show that our method accurately and efficiently re-ranks the neighbor hypotheses using little memory compared to the full vectors representation.

Affiliated with

Also associated with

No associations

Rating

Searching in one billion vectors: re-rank with source coding does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Searching in one billion vectors: re-rank with source coding, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Searching in one billion vectors: re-rank with source coding will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-212777