| 摘要 |
A system and method for spatiotemporal depth extraction of images are provided. The system and method provide for acquiring a sequence of images from a scene (502), the sequence including a plurality of successive frames of images, estimating the disparity of at least one point in a first image with at least one corresponding point in a second image for at least one frame (504,506), estimating motion of the at least one point in the first image (605), estimating the disparity of the at least one next successive frame based on the estimated disparity of at least one previous frame in a forward direction of the sequence (508), wherein the estimate disparity is compensated with the estimated motion, and minimizing the estimated disparity of each of the plurality of successive frames based on the estimated disparity of at least one previous frame in a backward direction of the sequence (512). |
