Condensing image databases when retrieval is based on non-metric distances

David W. Jacobs*, Daphna Weinshall, Yoram Gdalyahu

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

12 Scopus citations

Abstract

One of the key problems in appearance-based vision is understanding how to use a set of labeled images to classify new images. Classification systems that can model human performance often make use of similarity judgements that are non-metric. Existing condensing techniques for finding class representatives which are ill-suited to deal with non-metric dataspaces are presented. The importance of ideas learned from the experience of improving performance using both synthetic and real images are discussed.

Original languageAmerican English
Pages596-601
Number of pages6
StatePublished - 1998
Externally publishedYes
EventProceedings of the 1998 IEEE 6th International Conference on Computer Vision - Bombay, India
Duration: 4 Jan 19987 Jan 1998

Conference

ConferenceProceedings of the 1998 IEEE 6th International Conference on Computer Vision
CityBombay, India
Period4/01/987/01/98

Fingerprint

Dive into the research topics of 'Condensing image databases when retrieval is based on non-metric distances'. Together they form a unique fingerprint.

Cite this