COOPERATIVE INTEGRATION OF STEREOPSIS AND OPTIC FLOW COMPUTATION

A cooperative integration of stereopsis and optic flow computation is presented. Central to our approach is the modeling of the visual proce sses as a sequence of coupled Markov random fields by definition of su itable interprocess interactions based on some natural constraints. Th e integration makes each of the individual processes better constraine d and more reliable. Further, as a result of the integration, it becom es possible to obtain accurately the discontinuities in both the flow and the disparity fields along with the regions of stereo occlusion. R esults on both noisy synthetic image data and real images are presente d. (C) 1995 Optical Society of America