We present a new method for detecting point matches between two images. The main issue is how to preserve the global consistency of individual matches. Existing methods propagate local smoothness by relaxation or do combinatorial search for an optimal solution. Our method imposes non-local constraints that should be approximately satisfied across the image. We define the "confidence" of such "soft constraints" to all potential matches. The confidence is progressively updated by "mean-field approximation". Finally, the "hard" epipolar constraint is imposed by RANSAC. Using real images, we demonstrate that our method is robust to camera rotations and zooming changes.

Abstract. We present a novel approach to surface reconstruction from multiple images. The central idea is to explore the integration of both 3D stereo data and 2D calibrated images. This is motivated by the fact that only robust and accurate feature points that survived the geometry scrutiny of multiple images are reconstructed in space. The density insufficiency and the inevitable holes in the stereo data should be filled in by using information from multiple images. The idea is therefore to first construct small surface patches from stereo points, then to progressively propagate only reliable patches in their neighborhood from images into the whole surface using a best-first strategy. The problem reduces to searching for an optimal local surface patch going through a given set of stereo points from images. This constrained optimization for a surface patch could be handled by a local graph-cut that we develop. Real experiments demonstrate the usability and accuracy of the approach. 1

Abstract—We address the fundamental problem of matching in two static images. The remaining challenges are related to occlusion and lack of texture. Our approach addresses these difficulties within a perceptual organization framework, considering both binocular and monocular cues. Initially, matching candidates for all pixels are generated by a combination of matching techniques. The matching candidates are then embedded in disparity space, where perceptual organization takes place in 3D neighborhoods and, thus, does not suffer from problems associated with scanline or image neighborhoods. The assumption is that correct matches produce salient, coherent surfaces, while wrong ones do not. Matching candidates that are consistent with the surfaces are kept and grouped into smooth layers. Thus, we achieve surface segmentation based on geometric and not photometric properties. Surface overextensions, which are due to occlusion, can be corrected by removing matches whose projections are not consistent in color with their neighbors of the same surface in both images. Finally, the projections of the refined surfaces on both images are used to obtain disparity hypotheses for unmatched pixels. The final disparities are selected after a second tensor voting stage, during which information is propagated from more reliable pixels to less reliable ones. We present results on widely used benchmark stereo pairs. Index Terms—Stereo, occlusion, pixel correspondence, computer vision, perceptual organization, tensor voting. 1

Abstract not found

We address the problem of multiple view stereo from a perceptual organization perspective. Currently, the leading methods in the field are volumetric. They operate at the level of scene voxels and image pixels, without considering the structures depicted in them. On the other hand, many perceptual organization methods for binocular stereo are not extensible to more images. We present an approach where feature matching and structure reconstruction are addressed within the same framework. In order to handle noise, lack of image features, and discontinuities, we adopt a tensor representation for the data and tensor voting for information propagation. The key contributions of this paper are twofold. First, we introduce "saliency" instead of correlation as the criterion to determine the correctness of matches; second, our tensor representation and voting enable us to perform the complex computations associated with multiple view stereo at a reasonable computational cost. We present results on real data.

Abstract: This paper proposes a voting-based model that predicts depth at weakly-structured image areas from the depth that is extracted using a feature-based stereo method. We provide results, on both real and artificial scenes, that show the accuracy and robustness of our approach. Moreover, we compare our method to different dense stereo algorithms to investigate the effect of texture on performance of the two different approaches. The results confirm the expectation that dense stereo methods are suited better for textured image areas and our method for weakly-textured image areas. 1