Big Little Icons

Computer icons are small artificial images designed to be perceived with minimal ambiguity by the human visual system. In order to make them easier to perceive by visually impaired people, we propose a solution to the superresolution problem for color bitmap icons in a manner that exploits the unique characteristics of this medium versus that of generic low resolution natural imagery. We propose an MRF-based solution that incorporates local models of luminance and color perception which lays the basis for a snake-based vectorization of the icon and demonstrates encouraging performance on a diverse set of icons.