Relevance Feedback in CBIR [chapter]

Hongjiang Zhang, Zhong Su
2002 Visual and Multimedia Information Management  
A new focus in content-based image retrieval (CBIR) research is applying relevance feedback originally developed for text document retrieval, to improve the retrieval performance. This effort tries to bridge the gap between low-level image features and high-level semantic contents of images as this gap is the bottleneck of CBIR. We consider relevance feedback in CBIR a small sample-learning process in sparse image feature space. Almost all of the previously proposed methods fall well into such
more » ... ramework. In this paper, we present a brief overview of the current stateof-the-art of this topic, and present a framework of integrated relevance feedback and semantic learning in CBIR.
doi:10.1007/978-0-387-35592-4_3 fatcat:s56mbelgvzgvfcxhjlbnldkwma