Research project

The segmentation of skin regions in images is an important pre- processing step for many computer vision applications like face detection, face tracking or image retrieval systems. However, skin processing using color information can be a challenging task as the appearance of the skin in images is affected by different factors such as illumination, environment and ethnicity. The aim of this project is the analysis of the interaction of light with human skin from computer vision's perspective. The objective is the adaptation of state-of-the-art skin reflectance models to better suit computer vision related applications. Therein, two major parts have to be addressed: physics-based skin detection and illumination estimation based on the detected skin. Both aspects are highly correlated, as changes in the spectrum of the reflected light can be either due to a different skin albedo, or due to a change in the illumination. Illumination compensation can be used to enhance skin detection algorithms. Conversely, one can use previously detected skin regions to estimate the illuminant.