一种利用MAP估计的双目运动体视算法

本文提出一种融合运动和体视信息的运动与结构参数的鲁棒估计算法，算法首先通过计算瞬时ＦＯＥ，实现对每个像素处位移的ＭＡＰ估计，并计算关联的置信度；然后利用位移估计，从两图像序列之一计算相对深度，导出能用于匹配两体视图像序列的视差先验概率分布，使体视对的两帧图像匹配，确定３Ｄ景物深度，算法可方便地估计每个像素处的视差，不需对表面作平滑假设。

An Algorithm for Binocular Motion Stereo Using Map Estimation

This paper proposes an alogrithm for fusing motion and stereo cues to get robust estimates of both motion and structure. Our algorithm can compute a MAP estimate of the displacement at each pixel and an associated confidence measure by calculating the instantaneous FOE. Using the displacement estimates we calculate a relative depth map from one of the image sequences. This depth map is then used to derive a prior probability distribution for disparity that is used in matching the two frames of stereo paris. This method is used to estimate the disparity at each pixel independently and does not need any assumptions about smoothness.