Showing papers in "Computer Vision and Image Understanding in 2014"

TL;DR: From this large set of various BG methods, a relevant experimental analysis is conducted to evaluate both their robustness and their practical performance in terms of processor/memory requirements.

TL;DR: A thorough experimental evaluation vouches that SHOT outperforms state-of-the-art local descriptors in experiments addressing descriptor matching for object recognition, 3D reconstruction and shape retrieval.

TL;DR: This work aims to initiate a rigorous and comprehensive review of RPCA-PCP based methods for testing and ranking existing algorithms for foreground detection and investigates how these methods are solved and if incremental algorithms and real-time implementations can be achieved.

TL;DR: In this paper, a multiview Hessian discriminative sparse coding (mHDSC) is proposed to integrate Hessian regularization with sparse coding for image annotation.

TL;DR: A more comprehensive comparison of fifteen best sketch-based 3D shape retrieval methods is performed by completing the evaluation of each method on both benchmarks.

TL;DR: This work presents TouchCut, a robust and efficient algorithm for segmenting image and video sequences with minimal user interaction, and describes such a case study, enabling users to selectively stylize video objects to create a hand-painted effect.

TL;DR: A novel Linearly Approximated Sparse Representation-based Classification (LASRC) algorithm that uses linear regression to perform sample selection for @?^1-minimization, thus harnessing the speed of least-squares and the robustness of sparse solutions such as SRC.

TL;DR: An algorithm based on a self-adaptive Gaussian mixture to model the background of a scene imaged by a static video camera that uses a dynamic learning rate with adaptation to global illumination to cope with sudden variations of scene illumination.

TL;DR: This paper reviews how different types of models have been used in the literature, then proceeds to define the models and analyze them theoretically, in terms of both their statistical and computational aspects, and performs extensive experimental comparison on the task of model fitting.

TL;DR: The purpose of this paper is to survey recent research on the computer analysis of human beauty and present results in human sciences and medicine pointing to a largely shared and data-driven perception of attractiveness, which is a rationale of computer beauty analysis.

TL;DR: In this article, a methodology of classifying hepatic (liver) lesions using multidimensional persistent homology, the matching metric (also called the bottleneck distance), and a support vector machine was presented.

TL;DR: A new performance metric addressing and unifying the qualitative and quantitative aspects of the performance measures is proposed, which has been tested on several activity recognition algorithms participating in the ICPR 2012 HARL competition.

TL;DR: The 3dSOBS+ algorithm, a newly designed approach for moving object detection based on a neural background model automatically generated by a self-organizing method, is proposed, that compares well with the state-of-the-art methods.

TL;DR: A formulation of a random walk in a directed hypergraph that serves as a basis to a semi-supervised image segmentation procedure that is configured as a machine learning problem, where a few sample pixels are used to estimate the labels of the unlabeled ones.

TL;DR: This work proposes a novel method that uses a posture-invariant shape space to model body shape variation combined with a skeleton-based deformation to model posture variation and demonstrates that higher fitting accuracy is achieved than when using a variant of the popular SCAPE model.

TL;DR: SnooperText is described, an original detector for textual information embedded in photos of building facades (such as names of stores, products and services) that was developed for the iTowns urban geographic information project and outperforms other published state-of-the-art text detection algorithms on standard image benchmarks.

TL;DR: This paper proposes a novel global spatio-temporal self-similarity measure to score saliency using the ideas of dictionary learning and sparse coding, and demonstrates that this approach performs competitively to the state of the art.

TL;DR: An image classification framework by leveraging the non-negative sparse coding, correlation constrained low rank and sparse matrix decomposition technique (CCLR-Sc+SPM), which achieves or outperforms the state-of-the-art results on several benchmarks.

TL;DR: This paper proposes a novel approach to addressing the issue of accurate video-based ball tracking in team sports by formulating the tracking in terms of deciding which player, if any, is in possession of the ball at any given time.

TL;DR: It is shown to be more accurate and to require less iterations in LO-RANSAC based estimation, than the current point based approaches that employ the affine relation to generate pointwise correspondences and then calculate the fundamental matrix from the pointwise relations.

TL;DR: A novel feature selection method using a sparse model that is more flexible in selecting the discriminating features as it is able to control the degree of sparseness and considering both global and local structures of data distribution makes the feature selection process more effective.

TL;DR: A kernel density estimation method which models background and foreground by exploiting textons to describe textures within small and low contrasted regions and is able to generalize over different scenarios and targets is presented.

TL;DR: It is shown through practical experiments that, with implementations on GPUs, multi-object segmentation and tracking using state-of-art MRF inference methods is feasible, despite the computational costs typically associated with such methods.

TL;DR: Evidence is presented showing the ChESS feature detector, designed to exclusively respond to chess-board vertices, superior robustness, accuracy, and efficiency in comparison to other commonly used detectors, both under simulation and in experimental 3D reconstruction of flat plate and cylindrical objects.

TL;DR: Experimental results indicate that the proposed approach provides consistently better accuracy than other related multiphase active contour algorithms using four different error metrics even under severe noise, intensity inhomogeneities, and partial volume effects in MRI imagery.

TL;DR: A novel approach which takes into account prior information about the location of possible pointing targets is formulated, which uses the Dempster-Shafer theory of evidence to fuse information from two different input streams: head pose, estimated by visually tracking the off-plane rotations of the face, and hand pointing orientation.

TL;DR: This paper constructs ImageWeb, a sparse graph consisting of all the images in the database, in which two images are connected if and only if one is ranked among the top of another’s initial search result, and uses HITS, a query-dependent algorithm to re-rank the images according to the affinity values.

TL;DR: It is demonstrated that the proposed approach keeps solving the assignment problem tractable, while taking into account how different assignments influence feature measurement, on four challenging multi-person datasets (indoor and outdoor), involving 3–5 overlapping cameras and up to 23 persons simultaneously.

TL;DR: This paper investigates a new class of estimators on digital shape boundaries based on integral invariants and provides both proofs of multigrid convergence of principal curvature estimators and a complete experimental evaluation of their performances.

TL;DR: Experimental results on simulated and real data show that the convex point-wise method and the nonconvex method outperform respectively current initialization and refinement methods in 3D reconstructed surface accuracy.