Abstract This paper proposes a new approach for superpixel segmentation. It is formulated as finding a rooted spanning forest of a graph with respect to some roots and a path-cost function. The underlying graph represents an image, the roots serve as seeds for segmentation, each pixel is connected to one seed via a path, the path-cost function measures both the color similarity and spatial closeness between two pixels via a path, and each tree in the spanning forest represents one superpixel. Originating from the evenly distributed seeds, the superpixels are guided by a path-cost function to grow uniformly and adaptively, the pixel-by-pixel growing continues until they cover the whole image. The number of superpixels is controlled by the number of seeds. The connectivity is maintained by region growing. Good performances are assured by connecting each pixel to the similar seed, which are dominated by the path-cost function. It is evaluated by both the superpixel benchmark and supervoxel benchmark. Its performance is ranked as the second among top performing state-of-the-art methods. Moreover, it is much faster than the other superpixel and supervoxel methods.

Phase unwrapping (PU) is one of the key process in reconstructing the digital elevation model (DEM) of a scene from its interferometric synthetic aperture radar (InSAR) data. Compared with traditional single-baseline PU, the multibaseline PU does not need to obey the phase continuity assumption, which can be applicable to reconstruct the DEM where topography varies drastically. However, the performance of the multibaseline PU is directly concerned with noise level. Contrarily, the single-baseline PU algorithm has good noise robustness, since it is based on the globe wrapped phase information, such as PU-max-flow (PUMA) algorithm. In order to improve the noise robustness of the multibaseline, in this paper, we extend single-baseline PUMA algorithm to multibaseline domain, referred to as multibaseline PUMA algorithm, which allows the unwrapping of multibaseline interferograms for the generation of DEM. The proposed algorithm does not need to obey the phase continuity assumption by taking the advantages of multibaseline diversity and improves the noise robustness by using the global wrapped information both from single- and multibaseline domain. The performance of the proposed algorithm is tested on simulated InSAR data experiments, which demonstrate the effectiveness and noise robustness of the proposed algorithm.

This letter presents a new method to generate superpixel lattice. Unlike the existing methods based on strip seaming, it iteratively refines an initial lattice to achieve an expected lattice. In each iteration, it adjusts local lattice step by step. At each step, a vertex and its four connecting edges are refined by finding the optimal horizontal and vertical edges. The optimal edges are found by dynamic programming technique. This method is tested against both remotely sensed images and standard data set dedicated to segmentation evaluation. The proposed method is better or competitive with the existing state-of-the-art superpixel lattice methods concerning both performance and efficiency.

The automatic extraction of line-networks from images is a well-known computer vision issue. Appearance and shape considerations have been deeply explored in the literature to improve accuracy in presence of occlusions, shadows, and a wide variety of irrelevant objects. However most existing works have ignored the structural aspect of the problem. We present an original method which provides structurally-coherent solutions. Contrary to the pixel-based and object-based methods, our result is a graph in which each node represents either a connection or an ending in the line-network. Based on stochastic geometry, we develop a new family of point processes consisting in sampling junction-points in the input image by using a Monte Carlo mechanism. The quality of a configuration is measured by a probability density which takes into account both image consistency and shape priors. Our experiments on a variety of problems illustrate the potential of our approach in terms of accuracy, flexibility and efficiency.

This paper presents two novel approaches for stereo matching. First, a bitwise algorithm for stereo matching is proposed. It represents the disparity of each pixel as a binary number, treats each bit separately and determines them step by step, each step determines one bit of the disparities and involves only a single graph cut computation. It reduces the computation complexity to (9(log2 n). Second, bilayer stereo matching for scenes consisting of foreground and background is proposed. Disparity field of foreground and background layers are recovered independently, and the final disparity field is composed using the disparity field of two layers. Discontinuity between foreground and background can be preserved successfully by this schema. Experimental results show that the combination of bitwise algorithm and bilayer stereo matching can support realtime applications such as desktop-based 3D visual communication.