Abstract : This paper presents a component of a content based image retrieval system dedicated to let a user define the indexing terms used later during retrieval. A user inputs a indexing term name, image examples and counter-examples of the term,and the system learns a model of the concept as well as a similarity measure for this term. The similarity measure is based on weights reflecting the importance of each low-level feature extracted from the images. The system computes these weights using a genetic algorithm. Rating a particular similarity measure is done by clustering the examples and counter-examples using these weights and computing the quality of the obtained clusters. Experiments are conducted and results are presented on a set of 600 images.
https://hal.inria.fr/hal-00953933 Contributor : Marie-Christine FauvetConnect in order to contact the contributor Submitted on : Monday, March 3, 2014 - 12:49:24 PM Last modification on : Sunday, June 26, 2022 - 4:59:02 AM Long-term archiving on: : Saturday, May 31, 2014 - 10:51:42 AM
Stéphane Bissol, Philippe Mulhem, yves Chiaramella. Dynamic Learning of Indexing Concepts for Home Image Retrieval. Content-Based Multimedia Indexing (CBMI2003), 2003, Rennes, France. pp.87--93. ⟨hal-00953933⟩