Learning Query-Specific Distance Functions for Large-Scale Web Image Search
Venue
IEEE Transactions on Multimedia, vol. 15 (2013), pp. 2022-2034
Publication Year
2013
Authors
Yushi Jing, Michele Covell, David Tsai, James M. Rehg
BibTeX
Abstract
Current Google image search adopts a hybrid search approach in which a text-based
query (e.g., "Paris landmarks") is used to retrieve a set of relevant images, which
are then refined by the user (e.g., by re-ranking the retrieved images based on
similarity to a selected example). We conjecture that given such hybrid image
search engines, learning per-query distance functions over image features can
improve the estimation of image similarity. We proposed scalable solutions to
learning query-specific distance functions by 1) adopting a simple large-margin
learning framework, 2) using the query-logs of a text-based image search engine to
train distance functions used in content-based systems. We evaluate the feasibility
and efficacy of our proposed system through comprehensive human evaluation, and
compare the results with the state-of-the-art image distance function used by
Google image search.
