PBIR: perception-based image retrieval-a system that can quickly capture subjective image query concepts

We describe the Perception-Based Image Retrieval (PBIR) system that we have built on our recently developed query-concept learning algorithms, MEGA and SVMActive. We show that MEGA and SVMActive can learn a complex image-query concept in a small number of user iterations (usually three to four) on a large, multi-category, high-dimensional image database.

[1]  Edward Y. Chang,et al.  Clustering for Approximate Similarity Search in High-Dimensional Spaces , 2002, IEEE Trans. Knowl. Data Eng..

[2]  E. Y. Chang,et al.  Toward perception-based image retrieval , 2000, 2000 Proceedings Workshop on Content-based Access of Image and Video Libraries.

[3]  Edward Y. Chang,et al.  Learning image query concepts via intelligent sampling , 2001, IEEE International Conference on Multimedia and Expo, 2001. ICME 2001..

[4]  Kriengkrai Porkaew,et al.  Query refinement for multimedia similarity retrieval in MARS , 1999, MULTIMEDIA '99.

[5]  Christos Faloutsos,et al.  FALCON: Feedback Adaptive Loop for Content-Based Retrieval , 2000, VLDB.