Search by example, i.e. finding images that are similar to a query image, is an essential function for modern image search engines. It is mostly seen that the image search results obtained based on texture and visual are poor because of irrelevant relation between seeing and textural features. So, to avoid this, we decided to construct a ranking model based on user clicks rather than ranking them based on shape, colour, texture or other visual contents. Visual feature and click-based features are applied simultaneously to design a ranking model. In this paper, we propose a novel ranking model in which the user requires to only click on one query image with minimum effort and images from the database are retrieved and re-ranked based on visual features.
