Showing papers in "Computer Vision and Image Understanding in 2011"

TL;DR: This survey focuses on approaches that aim on classification of full-body motions, such as kicking, punching, and waving, and categorizes them according to how they represent the spatial and temporal structure of actions.

TL;DR: A sequential 2 stage approach is taken for pose classification and view dependent facial expression classification to investigate the effects of yaw variations from frontal to profile views and the influence of pose on different facial expressions.

TL;DR: A method is presented for categorizing manipulated objects and human manipulation actions in context of each other, able to simultaneously segment and classify human hand actions, and detect and classify the objects involved in the action.

TL;DR: Compared to other anomalous video event detection approaches that analyze object trajectories only, this work proposes a context-aware method to detect anomalies that is computationally efficient and can infer complex rules.

TL;DR: A framework of a medical image analysis system for the brain tumor segmentation and the brain tumors following-up over time using multi-spectral MRI images is presented and the quantitative evaluations by comparing with experts' manual traces and with other approaches demonstrate the effectiveness of the proposed method.

TL;DR: Wang et al. as discussed by the authors proposed a fully automatic new approach for color texture image segmentation based on neutrosophic set (NS) and multiresolution wavelet transformation, which aims to segment the natural scene images, in which the color and texture of each region does not have uniform statistical characteristics.

TL;DR: Combinatorial Multigrid (CMG) is presented, the first reliably efficient SDD solver that tackles problems in general and arbitrary weighted topologies and is validated on very large systems derived from imaging applications.

TL;DR: A framework which discovers location-representative tags from travelogues and then select relevant and representative photos to visualize these tags is proposed, which provides an informative summary which describes a given destination both textually and visually.

TL;DR: A generalized platform for constructing local shape descriptors that subsumes a large class of existing methods, and that allows for tuning to the geometry of specific models is presented.

TL;DR: This work synergistically combine two state-of-the-art methodologies, the ability to track and label single person trajectories in a crowded area using multiple video cameras, and a new class of novelty detection algorithms based on spectral analysis of graphs.

TL;DR: This contribution aims to review the state-of-the-art image registration methods that lay their foundations on evolutionary computation and aims to analyze the performance of some of the latter approaches when tackle a challenging real-world application in forensic anthropology, the 3D modeling of forensic objects.

TL;DR: The proposed composite splitting algorithms are applied to the compressed MR image reconstruction and low-rank tensor completion and demonstrate the superior performance of the proposed algorithms in terms of both accuracy and computation complexity.

TL;DR: A head-mounted gaze tracking system for the study of visual behavior in unconstrained environments designed both for adults and for infants as young as 1year of age that reaches an accuracy of 1.59^o with adults and 2.42^O with children.

TL;DR: The proposed approaches for recognizing human gestures from videos using models that are built from the Riemannian geometry of shape spaces are successfully able to represent the shape and dynamics of the different classes for recognition, but are also robust against some errors resulting from segmentation and background subtraction.

TL;DR: A new population game dynamics (InImDyn) is proposed which is motivated by the analogy with infection and immunization processes within a population of ''players,'' and it is proved that the evolution of the dynamics is governed by a quadratic Lyapunov function, representing the average population payoff.

TL;DR: Recognizing the occlusion by hair as one of the main obstacles hindering the deployment of ear biometrics, this work has specifically chosen its techniques to provide performance advantages in Occlusion.

TL;DR: A particle filter-based tracker that exploits a first order dynamic model and continuously performs adaptation of model noise so to balance uncertainty between the static and dynamic components of the state vector is proposed.

TL;DR: The main properties of the proposed registration method are noise robustness, outlier resistance and global optimal alignment, and a new stochastic approach for global minimization.

TL;DR: The ability to handle several thousand features per map at video-rate, and for the cameras to switch automatically between maps, allows spatially localized AR workcells to be constructed and used with very little intervention from the user of a wearable vision system.

TL;DR: In this article, a non-parametric combined segmentation and registration method is presented, where the shapes to be registered are implicitly modeled with level set functions and the problem is cast as an optimization one, combining a matching criterion based on the active contours without edges for segmentation (Chan and Vese, 2001) and a nonlinear-elasticity-based smoother on the displacement vector field.

TL;DR: The bootstrap method is based on bootstrap resampling, which is a computational statistical inference technique based on repeating the optical flow calculation several times for different randomly chosen subsets of pixel contributions to obtain useful estimates of geometrical and angular errors.

TL;DR: A novel method for calibrating display-camera setups from reflections in a user's eyes, a mirroring device that is always available, and a non-linear optimization strategy that exploits geometry constraints within the system to considerably improve the initial estimate are described.

TL;DR: This work proposes a novel shape parsing approach which is based on identifying and regularizing the ligature structure of a medial axis, leading to a bone graph, a medial abstraction which captures a more stable notion of an object's parts.

TL;DR: Two new methodologies to compute a sub-optimal common labelling are presented, focusing in extending the Graduated Assignment algorithm, although the methodology could be applied to other probabilistic graph-matching algorithms.

TL;DR: This paper proposes a general algorithm for driving the segmentation that uses the ground truth and current segmentation error to automatically simulate user interactions, and investigates four strategies for selecting which pixels will form the next interaction.

TL;DR: This intrinsic approach provides a robust and efficient method to directly study image processing, in particular, total variation problems on surfaces without requiring any preprocessing.

TL;DR: A new method for iris recognition based on elastic graph matching and Gabor wavelets is presented, using the circular Hough transform to determine the iris boundaries and it is found that, the elasticgraph matching is an effective matching performance.

TL;DR: As the extensive experimental results with various benchmark sequences demonstrate, the proposed algorithm outperforms, quantitatively and qualitatively, many other appearance-based approaches as well as methods using Gaussian Mixture Model (GMM), especially under sudden and drastic changes in illumination.

TL;DR: A new tracking method with improved efficiency and accuracy based on the subspace representation and particle filter, and a more physically meaningful proposal distribution of the particle filter with consideration of the nature of motion is presented.

TL;DR: Qualitative and quantitative comparisons of the results of the proposed cartilage segmentation method with semi-automatic segmentation methods demonstrate the potential of the proposal for clinical application.