Search

Depth information has been discovered to improve the performance of computer vision tasks, such as semantic segmentation and object recognition. However, careful acquisition of depth data needs highly developed depth sensors which are expensive. As a classic computer vision task, depth estimation from a single image has obtained promising results based on supervised learning methods. In this paper, we investigate the extension of color images with corresponding deep-regressed depth images in boosting the performance of semantic segmentation. Furthermore, the usage of combining color channels with the estimated depth or the ground truth depth channel is compared. Specifically, there are two stages in our work. Firstly, we adopt the framework of convolutional neural networks (CNN) for the depth estimation by combing the global depth network and the depth gradient network. After refining based on these two networks, the depth image map can be estimated in a deep-regressed manner. Secondly, after augmenting the color images with the predicted depth images, fully convolutional networks (FCN) are further used to implement the pixel-level semantic labeling. In the experiments, we employ two popular RGBD datasets, i.e., SUNRGBD and NYUDv2, for 37 and 40-class semantic segmentation, respectively. By comparing with the ground truth depth images, experimental results demonstrate that the networks trained on the estimated depth images can achieve comparable performance on facilitating the accuracy of semantic segmentation task. [ABSTRACT FROM AUTHOR]

Introduced in the mid-1970s as an intermediate step in proving a long-standing conjecture on arithmetic progressions, Szemerédi’s regularity lemma has emerged over time as a fundamental tool in different branches of graph theory, combinatorics and theoretical computer science. Roughly, it states that every graph can be approximated by the union of a small number of random-like bipartite graphs called regular pairs. In other words, the result provides us a way to obtain a good description of a large graph using a small amount of data, and can be regarded as a manifestation of the all-pervading dichotomy between structure and randomness. In this paper we will provide an overview of the regularity lemma and its algorithmic aspects, and will discuss its relevance in the context of pattern recognition research. [ABSTRACT FROM AUTHOR]

In this paper, we propose a spectral graph wavelet approach for 3D shape retrieval using the bag-of-features paradigm. In an effort to capture both local and global characteristics of a 3D shape, we present a three-step feature description framework. Local descriptors are first extracted via the spectral graph wavelet transform having the Mexican hat wavelet as a generating kernel. Then, mid-level features are obtained by embedding local descriptors into the visual vocabulary space using the soft-assignment coding step of the bag-of-features model. A global descriptor is subsequently constructed by aggregating mid-level features weighted by a geodesic exponential kernel, resulting in a matrix representation that describes the frequency of appearance of nearby codewords in the vocabulary. Then, we compare the global descriptor of a query to all global descriptors of the shapes in the dataset using a dissimilarity measure and find the closest shape. Experimental results on two standard 3D shape benchmarks demonstrate the effectiveness of the proposed shape retrieval approach in comparison with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Hierarchies, such as the tree of shapes, are popular representations for image simplification and segmentation thanks to their multiscale structures. Selecting meaningful level lines (boundaries of shapes) yields to simplify image while preserving intact salient structures. Many image simplification and segmentation methods are driven by the optimization of an energy functional, for instance the celebrated Mumford–Shah functional. In this paper, we propose an efficient approach to hierarchical image simplification and segmentation based on the minimization of the piecewise-constant Mumford–Shah functional. This method conforms to the current trend that consists in producing hierarchical results rather than a unique partition. Contrary to classical approaches which compute optimal hierarchical segmentations from an input hierarchy of segmentations, we rely on the tree of shapes, a unique and well-defined representation equivalent to the image. Simply put, we compute for each level line of the image an attribute function that characterizes its persistence under the energy minimization. Then we stack the level lines from meaningless ones to salient ones through a saliency map based on extinction values defined on the tree-based shape space. Qualitative illustrations and quantitative evaluation on Weizmann segmentation evaluation database demonstrate the state-of-the-art performance of our method. [ABSTRACT FROM AUTHOR]

Shape clustering is a difficult visual task due to large intra-class variations and small inter-class variations induced by shape articulation, rotation, occlusion, etc . To tackle this problem, we attempt to leverage the complementary nature among features of different statistics ( e.g. , skeleton-based descriptors and contour-based descriptors) for robust clustering. In this paper, a similarity fusion framework based on spectral analysis is proposed. The proposed method, which we call co-spectral, is a spectral clustering algorithm. It has two inborn merits for shape clustering: (1) it can automatically make use of the complementarity of various shape similarities based on a co-training framework; (2) it does not require shape representations to be vectors. Co-spectral is evaluated on several popular shape benchmarks. The experimental results demonstrate that co-spectral outperforms the state-of-the-art algorithms by a large margin. [ABSTRACT FROM AUTHOR]

In this paper, we propose a method for classification of human actions using pose based features. We demonstrate that statistical information of key movements of actions can be utilized in designing an efficient input representation, using fuzzy membership functions. The ability of stacked auto encoder to learn the underlying features of input data is exploited to recognize human actions. The efficacy of the proposed approach is demonstrated on CMU MOCAP and Berkeley MHAD datasets. [ABSTRACT FROM AUTHOR]

Defining and computing distances between tree structures is a classical area of study in theoretical computer science, with practical applications in the areas of computational biology, information retrieval, text analysis, and many others. In this paper, we focus on rooted, unordered, uniquely-labelled trees such as taxonomies and other hierarchies. For trees as these, we introduce the intuitive concept of a ‘local move’ operation as an atomic edit of a tree. We then introduce SuMoTED, a new edit distance measure between such trees, defined as the minimal number of local moves required to convert one tree into another. We show how SuMoTED can be computed using a scalable algorithm with quadratic time complexity. Finally, we demonstrate its use on a collection of music genre taxonomies. [ABSTRACT FROM AUTHOR]

Community detection in a social network is a well-known problem that has been studied in computer science since early 2000. The algorithms available in the literature mainly follow two strategies, one, which allows a node to be a part of multiple communities with equal membership, and the second considers a disjoint partition of the whole network where a node belongs to only one community. In this paper, we proposed a novel community detection algorithm which identifies fuzzy-rough communities where a node can be a part of many groups with different memberships of their association. The algorithm runs on a new framework of social network representation based on fuzzy granular theory. A new index viz. normalized fuzzy mutual information, to quantify the goodness of detected communities is used. Experimental results on benchmark data show the superiority of the proposed algorithm compared to other well known methods, particularly when the network contains overlapping communities. [ABSTRACT FROM AUTHOR]