DocumentCode :
357029
Title :
Invariance in content-based retrieval
Author :
Smeulders, Arnold ; Gevers, Theo ; Geusebroek, Jan-Mark ; Worring, Marcel
Author_Institution :
ISIS, Amsterdam Univ., Netherlands
Volume :
2
fYear :
2000
fDate :
2000
Firstpage :
675
Abstract :
The paper discusses the role of invariance in content-based image retrieval, and discusses future directions. In the repertoire of images under consideration, a narrow domain is a set of images containing a limited variety in the object´s appearances. In a narrow domain, the recording circumstances typically are similar or standardized over the whole domain. On the other end of the spectrum is a broad domain with a large variety in recording circumstances and appearances. The sensory gap between the object in the world and the information captured from a recording causes unwanted variance in the description of an object. We list conditions of variance, and we review some of features which are invariant under these conditions
Keywords :
content-based retrieval; image retrieval; content-based image retrieval; invariance; narrow domain; recording circumstances; Content based retrieval; Cultural differences; Disk recording; Geometry; Humans; Image retrieval; Intersymbol interference; Layout; Pixel; Surface texture;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multimedia and Expo, 2000. ICME 2000. 2000 IEEE International Conference on
Conference_Location :
New York, NY
Print_ISBN :
0-7803-6536-4
Type :
conf
DOI :
10.1109/ICME.2000.871452
Filename :
871452
Link To Document :
بازگشت