A simple and effective still image retrieval paradigm

Considering a classic image-retrieval problem, a user submits a set of index images to a system and through repeated interactions, the goals are narrowed to the image(s) that satisfies the user. To this purpose, conventional content-based image retrieval (CBIR) paradigm make uses of image processing and computer-vision techniques, and tries to understand the visual(color, shape, texture etc) image content. We propose a novel paradigm to this problem from a totally different angle. It attempts to use the human's intuition capabilities. Instead of processing images, the system simply accumulates records of user feedback and recycles them in the form of collaborative filtering, just like a purchase recommendation system such as Amazo web. we use paradigm by a term "content-free" image retrieval (CFIR). We discuss many issues of visual retrieval, argue for the idea of CFIR, and present results of experiment. The experiment results show that the performance of CFIR improve image performance.