Changxin Gao

Pyramid-based multi-structure local binary pattern for texture classification

Recently, the local binary pattern (LBP) has been widely used in texture classification. The conventional LBP methods only describe micro structures of texture images, such as edges, corners, spots and so on, although many of them show a good performance on texture classification. This situation still could not be changed, even though the multiresolution analysis technique is used in methods of local binary pattern. In this paper, we investigate the drawback of conventional LBP operators in describing some textures that has the same small structures but differential large structures. And a multi-structure local binary pattern operator is achieved by executing the LBP method on different layers of image pyramid. The proposed method is simple yet efficient to extract not only the micro structures but also the macro structures of texture images. We demonstrate the performance of our method on the task of rotation invariant texture classification. The experimental results on Outex database show advantages of the proposed method.

Probabilistic class structure regularized sparse representation graph for semi-supervised hyperspectral image classification

Abstract Graph-based semi-supervised learning (SSL), which performs well in hyperspectral image classification with a small amount of labeled samples, has drawn a lot of attention in the past few years. The key step of graph-based SSL is to construct a good graph to represent original data structures. Among the existing graph construction methods, sparse representation (SR) based methods have shown impressive performance on graph-based SSL. However, most SR based methods fail to take into consideration the class structure of data. In SSL, we can obtain a probabilistic class structure, which implies the probabilistic relationship between each sample and each class, of the whole data by utilizing a small amount of labeled samples. Such class structure information can help SR model to yield a more discriminative coefficients, which motivates us to exploit this class structure information in order to learn a discriminative graph. In this paper, we present a discriminative graph construction method called probabilistic class structure regularized sparse representation (PCSSR) approach, by incorporating the class structure information into the SR model, PCSSR can learn a discriminative graph from the data. A class structure regularization is developed to make use of the probabilistic class structure, and therefore to improve the discriminability of the graph. We formulate our problem as a constrained sparsity minimization problem and solve it by the alternating direction method with adaptive penalty (ADMAP). The experimental results on Hyperion and AVIRIS hyperspectral data show that our method outperforms state of the art.

Text detection approach based on confidence map and context information

Abstract Text information plays a significant role in many applications for providing more descriptive and abstract information than other objects. In this paper, an approach based on the confidence map and context information is proposed to robustly detect texts in natural scenes. Most of the conventional methods design sophisticated texture features to describe the text regions, while we focus on building a confidence map model by integrating the seed candidate appearance and the relationships with its adjacent candidates to highlight the texts from the backgrounds, and the candidates with low confidence value will be removed. In order to improve the recall rate, the text context information is adopted to regain the missing text regions. Finally, the text lines are formed and further verified, and the words are obtained by calculating the threshold to separate the intra-word letters from the inter-word letters. Experimental results on the three public benchmark datasets, i.e., ICDAR 2005, ICDAR 2011 and ICDAR 2013, show that the proposed approach has achieved the competitive performances by comparing with the other state-of-the-art methods.

Robust Visual Tracking Using Exemplar-Based Detectors

Tracking by detection has become an attractive tracking technique, which treats tracking as an object detection problem and trains a detector to separate the target object from the background in each frame. While this strategy is effective to some extent, we argue that the task in tracking should be searching for a specific object instance instead of an object category. Based on this viewpoint, a novel framework based on object exemplar detectors is proposed for visual tracking. To build a specific and discriminative model to separate the object instance from the background, the proposed method trains an exemplar-based linear discriminant analysis (ELDA) classifier for the object exemplar, using the current tracked instance as the positive sample and massive negative samples obtained both offline and online. To improve the trackers’ adaptivity, we use an ensemble of the above ELDA detectors and update them during the tracking to cover the variation in object appearance. Extensive experimental results on a large benchmark data set show that the proposed method outperforms many state-of-the-art trackers, demonstrating the effectiveness and robustness of the ELDA tracker.

DeepList: Learning Deep Features With Adaptive Listwise Constraint for Person Reidentification

Person reidentification (re-id) aims to match a specific person across nonoverlapping cameras, which is an important but challenging task in video surveillance. Conventional methods mainly focus either on feature constructing or metric learning. Recently, some deep learning-based methods have been proposed to learn image features and similarity measures jointly. However, current deep models for person re-id are usually trained with either pairwise loss, where the number of negative pairs greatly outnumbering that of positive pairs may lead the training model to be biased toward negative pairs or constant margin hinge loss, without considering the fact that hard negative samples should be paid more attention in the training stage. In this paper, we propose to learn deep representations with an adaptive margin listwise loss. First, ranking lists instead of image pairs are used as training samples, in this way, the problem of data imbalance is relaxed. Second, by introducing an adaptive margin parameter in the listwise loss function, it can assign larger margins to harder negative samples, which can be interpreted as an implementation of the automatic hard negative mining strategy. To gain robustness against changes in poses and part occlusions, our architecture combines four convolutional neural networks, each of which embeds images from different scales or different body parts. The final combined model performs much better than each single model. The experimental results show that our approach achieves very promising results on the challenging CUHK03, CUHK01, and VIPeR data sets.

On selection and combination of weak learners in AdaBoost

Despite of its great success, two key problems are still unresolved for AdaBoost algorithms: how to select the most discriminative weak learners and how to optimally combine them. In this paper, a new AdaBoost algorithm is proposed to make improvement in the two aspects. First, we select the most discriminative weak learners by minimizing a novel distance related criterion, i.e., error-degree-weighted training error metric (ETEM) together with generalization capability metric (GCM), rather than training error rate only. Second, after getting the coefficients that are set empirically, we combine the weak learners optimally by tuning the coefficients using kernel-based perceptron. Experiments with synthetic and real scene data sets show our algorithm outperforms conventional AdaBoost.

A Computational Model for Object-Based Visual Saliency: Spreading Attention Along Gestalt Cues

The past few years have witnessed impressive progress on the research of salient object detection. Nevertheless , existing approaches still cannot perform satisfactorily in the case of complex scenes, particularly when the salient objects have non- uniform appearance or complicated shapes, and the background is complexly structured. One important reason for such limitations may be that these approaches commonly ignore the factor of perceptual grouping in saliency modeling. To address this issue, this paper presents a novel computational model for object -based visual saliency, which explicitly takes into consideration the connections between attention and perceptual grouping, and incorporates Gestalt grouping cues into saliency computation. Inspired by the sensory enhancement theory, we suggest a paradigm for object-based saliency modeling, that is, object-based saliency stems from spreading attention along Gestalt grouping cues. Computationally , three typical Gestalt cues, including proximity, similarity, and closure, are respectively extracted from the given image, which are then integrated by constructing a unified Gestalt graph. A new algorithm named personalized power iteration clustering is developed to effectively fulfill the spreading of attention information across the Gestalt graph. Intensive experiments have been carried out to demonstrate the superior performance of the proposed model in comparison to the state-of-the-art.

Similarity Learning with Top-heavy Ranking Loss for Person Re-identification

Person re-identification is the task of finding a person of interest across a network of cameras. In this paper, we propose a new similarity learning method for person re-identification. Conventional metric learning methods generally learn a linear transformation by employing sparse pairwise or triplet constraints. Since a lot of negative matching pairs or triplets are abandoned, the discriminative information is not fully exploited. Similarity learning methods with AUC loss can utilize all valid triplet constraints. However, the AUC loss has its own limitation by treating all false ranks occured at different positions equally. To address this limitation, we propose to extend the AUC loss to the top-heavy ranking loss by assigning large weights to top positions of the ranking list. Moreover, we introduce an explicit nonlinear transformation function for the original feature space and learn an inner product similarity under the structured output learning framework. Our approach achieves very promising results on the challenging VIPeR, CUHK Campus and PRID 450S datasets.

Multi-structure local binary patterns for texture classification

Recently, the local binary patterns (LBP) have been widely used in the texture classification. The LBP methods obtain the binary pattern by comparing the gray scales of pixels on a small circular region with the gray scale of their central pixel. The conventional LBP methods only describe microstructures of texture images, such as edges, corners, spots and so on, although many of them show good performances on the texture classification. This situation still could not be changed, even though the multi-resolution analysis technique is adopted by LBP methods. Moreover, the circular sampling region limits the ability of the conventional LBP methods in describing anisotropic features. In this paper, we change the shape of sampling region and get an extended LBP operator. And a multi-structure local binary pattern (Ms-LBP) operator is achieved by executing the extended LBP operator on different layers of an image pyramid. Thus, the proposed method is simple yet efficient to describe four types of structures: isotropic microstructure, isotropic macrostructure, anisotropic microstructure and anisotropic macrostructure. We demonstrate the performance of our method on two public texture databases: the Outex and the CUReT. The experimental results show the advantages of the proposed method.

Temporally aligned pooling representation for video-based person re-identification

This paper proposes an effective Temporally Aligned Pooling Representation (TAPR) for video-based person re-identification. To extract the motion information from a sequence, we propose to track the superpixels of the lowest portions of human. To perform temporal alignment of videos, we propose to select the “best” walking cycle from the noisy motion information according to the intrinsic periodicity property of walking persons, that is fitted sinusoid in our implementation. To describe the video data in the selected walking cycle, we first divide the cycle into several segments according to the sinusoid, and then describe each segment by temporally aligned pooling. Extensive experimental results on the public datasets demonstrate the effectiveness of the proposed method compared with the state-of-the-art approaches.