To improve upon the initial disparity estimates stemming from a local correspondence method, a subsequent refinement step is commonly employed. The performance of the stereo matcher used for the initial estimation inherently depends on the underlying image content. For some regions of an image, it may be difficult or even physically impossible to establish accurate point correspondences. This results in disparity estimates of varying accuracy and reliability. In this paper, a confidence map is proposed which combines the consistency and quality of a match. It explicitly models the reliability of each disparity estimate. Such a confidence map represents valuable, additional information that can be leveraged in subsequent steps of the 3D processing chain. In this regard, this paper presents an extension of a cross bilateral filter that leverages this reliability information during a fast-converging refinement step in order to create robust and reliable disparity maps.
A wide variety of computer vision applications rely on superpixel or supervoxel algorithms as a preprocessing step. This underlines the overall importance that these approaches have gained in recent years. However, most methods show a lack of temporal consistency or fail in producing temporally stable superpixels. In this paper, we present an approach to generate temporally consistent superpixels for video content. Our method is formulated as a contour-evolving expectation-maximization framework, which utilizes an efficient label propagation scheme to encourage the preservation of superpixel shapes and their relative positioning over time. By explicitly detecting the occlusion of superpixels and the disocclusion of new image regions, our framework is able to terminate and create superpixels whose corresponding image region becomes hidden or newly appears. Additionally, the occluded parts of superpixels are incorporated in the further optimization. This increases the compliance of the superpixel flow with the optical flow present in the scene. Using established benchmark suites, we show the performance of our approach in comparison to state-of-the-art supervoxel and superpixel algorithms for video content.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.