[CEO08] Speeding Up Active Relevance Feedback with Approximate kNN Retrieval for Hyperplane Queries

Revue Internationale avec comité de lecture : Journal International Journal of Imaging Systems and Technology, vol. 18(2), pp. 150-159, 2008, (doi:10.1002/ima.20151)
Résumé: In content-based image retrieval, relevance feedback is a prominent method for reducing the "semantic gap" between the low-level features describing the content and the usually higher-level meaning of user's target. Recent relevance feedback methods are able to identify complex target classes after relatively few feedback iterations. However, since the computational complexity of such methods is linear in the size of the database, retrieval can be quite slow on very large databases. To address this scalability issue for active learning-based relevance feedback, we put forward a method that consists in the construction of an index in the feature space associated to a kernel function and in performing approximate kNN hyperplane queries with this feature space index. The experimental evaluation performed on two image databases show that a significant speedup can be achieved, at the expense of a limited increase in the number of feedback rounds.


@article {
title="{Speeding Up Active Relevance Feedback with Approximate kNN Retrieval for Hyperplane Queries}",
author="M. Crucianu and D. Estevez and V. Oria and J. Tarel",
journal="International Journal of Imaging Systems and Technology",