Region of interest in images using tagged information

Abstract

With the rise in popularity of photo sharing websites, the concept of tagging has been introduced to provide visual descriptions of the images and to make them searchable. Besides that, different regions of an image can be indicated with different tags and this idea is being employed in the methodology covered in our project. This project proposes a software implementation that is able to take a query image and its tag to detect the region of interest (ROI) in it after comparing it with similarly tagged images in the database. A successful automatic ROI detection technique can then allow for the automatic tagging of large collections of images by comparing them with manually tagged image(s) which contains all possible tags.