INTEGRATION OF SHAPE FROM SHADING AND STEREO

Stereo algorithms suffer from the lack of local surface texture due to smoothness of depth constraint, or local miss-matches in disparity es timates. Thus, most stereo methods only provide a coarse depth map whi ch can be associated with a low pass image of the depth map. On the ot her hand, shape from shading algorithms generally produce better estim ates of local surface areas, but some of them have problems with varia ble albedo and spherical surfaces. Thus, shape from shading methods pr oduce better detailed depth information, and can be associated with th e high pass image of the depth map. In order to compute a better depth map, we present a method for integrating the high frequncy informatio n from the shape from shading and the low frequency information from s tereo. The proposed algorithm is very simple, takes about 0.7 s for a 128 x 128 image on a Sun SparcStation-1, is non-iterative, and require s very little adjustment of parameters. The results obtained with a va riety of synthetic and real images are discussed. The quality of depth obtained by integrating shading and stereo is compared with the groun d truth (range image) using height error measure, and improvement rang ing from 30 to 50% over stereo, and from 65 to 98% over shading is dem onstrated.