In computer vision a large number of problems is about finding boundaries in a scene which are usually ill-defined due to lack of resolution, noise and occlusions, etc. Traditional approaches such as regularization and the well-known Laplacian of Gaussian (LoG) type filters throughout the 70s to early 90s have not led to satisfactory results. We have found, however, combining with unsupervised learning, more specifically, clustering, with the theory of Markov Random Field (MRF) we are able to achieve marked improvements over some of the major techniques recently reported in the literature. In this talk I will address in particular the problem of self-validated labeling of Markov random fields (MRFs), namely to optimize an MRF with unknown number of labels.1 I will present graduated graph cuts (GGC), a new technique that extends the binary s-t graph cut for self-validated labeling. Specifically, we use the split-and merge strategy to decompose the complex problem to a series of tractable subproblems. In terms of Gibbs energy minimization, a suboptimal labeling is gradually obtained based upon a set of cluster-level operations. By using different optimization structures, we propose three practical algorithms: tree-structured graph cuts (TSGC), net-structured graph cuts (NSGC) and hierarchical graph cuts (HGC). In contrast to previous methods, the proposed algorithms can automatically determine the number of labels, properly balance the labeling accuracy, spatial coherence and the labeling cost (i.e., the number of labels), and are computationally efficient, independent to initialization and able to converge to good local minima. We apply the proposed algorithms to natural image segmentation. Experimental results show that our algorithms produce generally feasible segmentations for Benchmark datasets, and outperform alternative methods in terms of robustness to noise, speed and preservation of soft boundaries.
展开▼
