Learning to Deblur Images with Exemplars.

Human faces are one interesting object class with numerous applications. While significant progress has been made in the generic deblurring problem, existing methods are less effective for blurry face images. The success of the state-of-the-art image deblurring algorithms stems mainly from implicit or explicit restoration of salient edges for kernel estimation. However, existing methods are less effective as only few edges can be restored from blurry face images for kernel estimation. In this paper, we address the problem of deblurring face images by exploiting facial structures. We propose a deblurring algorithm based on an exemplar dataset without using coarse-to-fine strategies or heuristic edge selections. In addition, we develop a convolutional neural network to restore sharp edges from blurry images for deblurring. Extensive experiments against the state-of-the-art methods demonstrate the effectiveness of the proposed algorithm for deblurring face images. In addition, we show that the proposed algorithms can be applied to image deblurring for other object classes.

Exploiting Spatial-Temporal Locality of Tracking via Structured Dictionary Learning.

In this paper, a novel spatial-temporal locality is proposed and unified via a discriminative dictionary learning framework for visual tracking. By exploring the strong local correlations between temporally obtained target and their spatially distributed nearby background neighbors, a spatial-temporal locality is obtained. The locality is formulated as a subspace model and exploited under a unified structure of discriminative dictionary learning with a subspace structure. Using the learned dictionary, the target and its background can be described and distinguished effectively through their sparse codes. As a result, the target is localized by integrating both the descriptive and the discriminative qualities. Extensive experiments on various challenging video sequences demonstrate the superior performance of proposed algorithm over the other state-of-the-art approaches.

Motion Blur Kernel Estimation via Deep Learning.

The success of the state-of-the-art deblurring methods mainly depends on the restoration of sharp edges in a coarse-to-fine kernel estimation process. In this paper, we propose to learn a deep convolutional neural network for extracting sharp edges from blurred images. Motivated by the success of the existing filtering-based deblurring methods, the proposed model consists of two stages: suppressing extraneous details and enhancing sharp edges. We show that the two-stage model simplifies the learning process and effectively restores sharp edges. Facilitated by the learned sharp edges, the proposed deblurring algorithm does not require any coarse-to-fine strategy or edge selection, thereby significantly simplifying kernel estimation and reducing computation load. Extensive experimental results on challenging blurry images demonstrate that the proposed algorithm performs favorably against the state-of-the-art methods on both synthetic and real-world images in terms of visual quality and run-time.

An Efficient Globally Optimal Algorithm for Asymmetric Point Matching.

Although the robust point matching algorithm has been demonstrated to be effective for non-rigid registration, there are several issues with the adopted deterministic annealing optimization technique. First, it is not globally optimal and regularization on the spatial transformation is needed for good matching results. Second, it tends to align the mass centers of two point sets. To address these issues, we propose a globally optimal algorithm for the robust point matching problem in the case that each model point has a counterpart in scene set. By eliminating the transformation variables, we show that the original matching problem is reduced to a concave quadratic assignment problem where the objective function has a low rank Hessian matrix. This facilitates the use of large scale global optimization techniques. We propose a modified normal rectangular branch-and-bound algorithm to solve the resulting problem where multiple rectangles are simultaneously subdivided to increase the chance of shrinking the rectangle containing the global optimal solution. In addition, we present an efficient lower bounding scheme which has a linear assignment formulation and can be efficiently solved. Extensive experiments on synthetic and real datasets demonstrate the proposed algorithm performs favorably against the state-of-the-art methods in terms of robustness to outliers, matching accuracy, and run-time.

Co-Bootstrapping Saliency.

In this paper, we propose a visual saliency detection algorithm to explore the fusion of various saliency models in a manner of bootstrap learning. First, an original bootstrapping model, which combines both weak and strong saliency models, is constructed. In this model, image priors are exploited to generate an original weak saliency model, which provides training samples for a strong model. Then, a strong classifier is learned based on the samples extracted from the weak model. We use this classifier to classify all the salient and non-salient superpixels in an input image. To further improve the detection performance, multi-scale saliency maps of weak and strong model are integrated, respectively. The final result is the combination of the weak and strong saliency maps. The original model indicates that the overall performance of the proposed algorithm is largely affected by the quality of weak saliency model. Therefore, we propose a co-bootstrapping mechanism, which integrates the advantages of different saliency methods to construct the weak saliency model thus addresses the problem and achieves a better performance. Extensive experiments on benchmark data sets demonstrate that the proposed algorithm outperforms the state-of-the-art methods.

Robust Visual Tracking via Convolutional Networks Without Training.

Deep networks have been successfully applied to visual tracking by learning a generic representation offline from numerous training images. However, the offline training is time-consuming and the learned generic representation may be less discriminative for tracking specific objects. In this paper, we present that, even without offline training with a large amount of auxiliary data, simple two-layer convolutional networks can be powerful enough to learn robust representations for visual tracking. In the first frame, we extract a set of normalized patches from the target region as fixed filters, which integrate a series of adaptive contextual filters surrounding the target to define a set of feature maps in the subsequent frames. These maps measure similarities between each filter and useful local intensity patterns across the target, thereby encoding its local structural information. Furthermore, all the maps together form a global representation, via which the inner geometric layout of the target is also preserved. A simple soft shrinkage method that suppresses noisy values below an adaptive threshold is employed to de-noise the global representation. Our convolutional networks have a lightweight structure and perform favorably against several state-of-the-art methods on the recent tracking benchmark data set with 50 challenging videos.

Image Deblurring via Enhanced Low-Rank Prior.

Low-rank matrix approximation has been successfully applied to numerous vision problems in recent years. In this paper, we propose a novel low-rank prior for blind image deblurring. Our key observation is that directly applying a simple low-rank model to a blurry input image significantly reduces the blur even without using any kernel information, while preserving important edge information. The same model can be used to reduce blur in the gradient map of a blurry input. Based on these properties, we introduce an enhanced prior for image deblurring by combining the low rank prior of similar patches from both the blurry image and its gradient map. We employ a weighted nuclear norm minimization method to further enhance the effectiveness of low-rank prior for image deblurring, by retaining the dominant edges and eliminating fine texture and slight edges in intermediate images, allowing for better kernel estimation. In addition, we evaluate the proposed enhanced low-rank prior for both the uniform and the non-uniform deblurring. Quantitative and qualitative experimental evaluations demonstrate that the proposed algorithm performs favorably against the state-of-the-art deblurring methods.

Robust superpixel tracking.

While numerous algorithms have been proposed for object tracking with demonstrated success, it remains a challenging problem for a tracker to handle large appearance change due to factors such as scale, motion, shape deformation, and occlusion. One of the main reasons is the lack of effective image representation schemes to account for appearance variation. Most of the trackers use high-level appearance structure or low-level cues for representing and matching target objects. In this paper, we propose a tracking method from the perspective of midlevel vision with structural information captured in superpixels. We present a discriminative appearance model based on superpixels, thereby facilitating a tracker to distinguish the target and the background with midlevel cues. The tracking task is then formulated by computing a target-background confidence map, and obtaining the best candidate by maximum a posterior estimate. Experimental results demonstrate that our tracker is able to handle heavy occlusion and recover from drifts. In conjunction with online update, the proposed algorithm is shown to perform favorably against existing methods for object tracking. Furthermore, the proposed algorithm facilitates foreground and background segmentation during tracking.