UCSC-SOE-11-18: Selective Removal of Extraneous Photographs

K. Armin Samii, Uliana Popov, Allison Carlisle, James Davis
07/12/2011 09:00 AM
Computer Science
In this paper, we propose a method for selecting the most representative high-quality images from a set of user photographs. To avoid redundancy arising from many similar images, we find all sets of near-duplicates. We then rank images based on three technical qualities: exposure, blur, and contrast. Finally, we provide an ordering of the images which accounts for both quality and uniqueness. Our goal is not to rank images based on subjective aesthetic qualities, but instead to help a photographer filter out technically flawed photographs and focus on objectively high quality shots.