Affiliations: [a] Department of Electrical Engineering, The City College of New York, New York, NY, USA | [b] Arlene R Gordon Research Institute, Lighthouse International, New York, NY, USA
Abstract: Matching clothes is a challenging task for many blind people. In this paper, we present a proof of concept system to solve this problem. The system consists of 1) a camera connected to a computer to perform pattern and color matching process; 2) speech commands for system control and configuration; and 3) audio feedback to provide matching results for both color and patterns of clothes. This system can handle clothes without any pattern, as well as clothing with multiple colors and complex patterns to aid both blind and color deficient people. Furthermore, our method is robust to variations of illumination, clothing rotation and wrinkling. To evaluate the proposed prototype, we collect two challenging databases including clothes without any pattern, or with multiple colors and different patterns under different conditions of lighting and rotation. Results reported here demonstrate the robustness and effectiveness of the proposed clothing matching system.
Keywords: Blind, color blind, computer vision, clothing matching, color matching, pattern analysis, visually impaired
