Qingming Leng

Zero-Shot Person Re-identification via Cross-View Consistency

Person re-identification, aiming to identify images of the same person from various cameras configured in different places, has attracted much attention in the multimedia retrieval community. In this problem, choosing a proper distance metric is a crucial aspect, and many classic methods utilize a uniform learnt metric. However, their performance is limited due to ignoring the zero-shot and fine-grained characteristics presented in real person re-identification applications. In this paper, we investigate two consistencies across two cameras, which are cross-view support consistency and cross-view projection consistency. The philosophy behind it is that, in spite of visual changes in two images of the same person under two camera views, the support sets in their respective views are highly consistent, and after being projected to the same view, their context sets are also highly consistent. Based on the above phenomena, we propose a data-driven distance metric (DDDM) method, re-exploiting the training data to adjust the metric for each query-gallery pair. Experiments conducted on three public data sets have validated the effectiveness of the proposed method, with a significant improvement over three baseline metric learning methods. In particular, on the public VIPeR dataset, the proposed method achieves an accuracy rate of 42.09% at rank-1, which outperforms the state-of-the-art methods by 4.29%.

Person re-identification with content and context re-ranking

This paper proposes a novel and efficient re-ranking technque to solve the person re-identification problem in the surveillance application. Previous methods treat person re-identification as a special object retrieval problem, and compute the retrieval result purely based on a unidirectional matching between the probe and all gallery images. However, the correct matching may be not included in the top-k ranking result due to appearance changes caused by variations in illumination, pose, viewpoint and occlusion. To obtain more accurate re-identification results, we propose to reversely query every gallery person image in a new gallery composed of the original probe person image and other gallery person images, and revise the initial query result according to bidirectional ranking lists. The behind philosophy of our method is that images of the same person should not only have similar visual content, refer to content similarity, but also possess similar k-nearest neighbors, refer to context similarity. Furthermore, the proposed bidirectional re-ranking method can be divided into offline and online parts, where the majority of computation load is accomplished by the offline part and the online computation complexity is only proportional to the size of the gallery data set, which is especially suited to the real-time required video investigation task. Extensive experiments conducted on a series of standard data sets have validated the effectiveness and efficiency of our proposed method.

Person Reidentification via Ranking Aggregation of Similarity Pulling and Dissimilarity Pushing

Person reidentification is a key technique to match different persons observed in nonoverlapping camera views. Many researchers treat it as a special object-retrieval problem, where ranking optimization plays an important role. Existing ranking optimization methods mainly utilize the similarity relationship between the probe and gallery images to optimize the original ranking list, but seldom consider the important dissimilarity relationship. In this paper, we propose to use both similarity and dissimilarity cues in a ranking optimization framework for person reidentification. Its core idea is that the true match should not only be similar to those strongly similar galleries of the probe, but also be dissimilar to those strongly dissimilar galleries of the probe. Furthermore, motivated by the philosophy of multiview verification, a ranking aggregation algorithm is proposed to enhance the detection of similarity and dissimilarity based on the following assumption: the true match should be similar to the probe in different baseline methods. In other words, if a gallery blue image is strongly similar to the probe in one method, while simultaneously strongly dissimilar to the probe in another method, it will probably be a wrong match of the probe. Extensive experiments conducted on public benchmark datasets and comparisons with different baseline methods have shown the great superiority of the proposed ranking optimization method.

Camera Compensation Using a Feature Projection Matrix for Person Reidentification

Matching individuals within a group of spatially nonoverlapping surveillance cameras, also known as person reidentification, has recently attracted a lot of research interest. Current methods mainly focus on feature representation or distance measure, which directly compare person images captured by different cameras. However, it is still a problem because of various surveillance conditions; for example, view switching, lighting variations, and image scaling. Although the brightness transfer function was proposed to address the problem of illumination variation, it could not handle view and scale changes among various cameras. In this paper, we propose a new approach to compensate for the inconsistency of feature distributions of person images captured by different cameras. More precisely, a feature projection matrix (FPM) is learned to project image features of one camera to the feature space of another camera, from which the latent device difference can be effectively eliminated for the person reidentification task. In particular, we formulate the FPM learning as a smooth unconstrained convex optimization problem and use a simple gradient descent algorithm with stochastic samples to accelerate the solving process. Extensive comparative experiments conducted on three standard datasets have shown the promising prospect of the proposed method.

Region-Based Interactive Ranking Optimization for Person Re-identification

Person re-identification, aiming to identify images of the same person from various cameras configured in difference places, has attracted plenty of attention in the multimedia community. Previous work mainly focuses on feature presentation and distance measure, and achieves promising results on some standard databases. However, the performance is still not good enough due to appearance changes caused by variations in illuminations, poses, viewpoints and occlusion. This paper addresses the problem through result re-ranking by introducing user feedback. In particular, considering the peculiarity of scarce positive and global similar negative samples in the person re-identification problem, we propose a region-based interactive ranking optimization method, to improve the original query result by labeling locally similar and dissimilar image regions. Experiments conducted on two standard data sets have validated the effectiveness of the proposed method with an average improvement of 10-30% over original basic method. It is proved that the ranking optimization algorithm is both an effective and efficient method to improve the original person re-identification result.

Bidirectional ranking for person re-identification

This paper proposes a simple but efficient bidirectional ranking method to improve person re-identification results across non-overlapping cameras. Previous methods treat person reidentification as a special object retrieval problem, and compute the final rank result purely based on a unidirectional matching between the probe and all gallery images. However, the expected person image may be excluded from the probes ??-nearest neighbor due to appearance changes caused by variations in illuminations, poses, viewpoints and occlusion. To solve the above problem, our method queries every gallery image in a new gallery composed of the original probe image and other gallery images, and revises the initial query result in accordance with both content and context similarities between bidirectional ranking lists. A latent assumption of our method is that images of the same person should not only have similar visual content, known as content similarity, but also possess similar k-nearest neighbors, known as context similarity. Extensive experiments conducted on a series of standard data sets have validated the effectiveness of our proposed method with an average improvement of 5-10% over original baseline methods.

Ranking Optimization for Person Re-identification via Similarity and Dissimilarity

Person re-identification is a key technique to match different persons observed in non-overlapping camera views.Many researchers treat it as a special object retrieval problem, where ranking optimization plays an important role. Existing ranking optimization methods utilize the similarity relationship between the probe and gallery images to optimize the original ranking list in which dissimilarity relationship is seldomly investigated. In this paper, we propose to use both similarity and dissimilarity cues in a ranking optimization framework for person re-identification. Its core idea is based on the phenomenon that the true match should not only be similar to the strong similar samples of the probe but also dissimilar to the strong dissimilar samples. Extensive experiments have shown the great superiority of the proposed ranking optimization method.

Coupled-View Based Ranking Optimization for Person Re-identification

Person re-identification aims to match different persons observed in non-overlapping camera views. Researchers have proposed many person descriptors based on global or local descriptions, while both of them have achieved satisfying matching results, however, their ranking lists usually vary a lot for the same query person. These motivate us to investigate an approach to aggregate them to optimize the original matching results. In this paper, we proposed a coupled-view based ranking optimization method through cross KNN rank aggregation and graph-based re-ranking to revise the original ranking lists. Its core assumption is that the images of the same person should share the similar visual appearance in both global and local views. Extensive experiments on two datasets show the superiority of our proposed method with an average improvement of 20-30% over the state-of-the-art methods at CMC@1.

Camera compensation using feature projection matrix for person re-identification

Matching individuals across a group of spatially non-overlapping surveillance cameras, also known as person re-identification, has recently attracted a lot of research interests. Current methods mainly focus on feature extraction or metric learning, which directly compare person images captured by different cameras, but seldom consider device differences caused by various surveillance conditions, e.g. view switching, scale zooming and illumination variation. Although brightness transfer function was proposed to address the problem of illumination variation, it could not handle view and scale changes among various cameras. In this paper, we propose an effective data-driven method to conquer device differences in the practical surveillance camera network. More precisely, with the help of a set of labelled pair-wise person images captured by two disjoint cameras, a feature projection matrix can be learned to project the person images of one camera to the feature space of the other camera, and thus images from these two different cameras can be accurately compared in a common feature space. Extensive comparative experiments conducted on three standard datasets have shown the promising prospect of our proposed methods.

Person Re-identification Using Data-Driven Metric Adaptation

Person re-identification, aiming to identify images of the same person from various cameras configured in difference places, has attracted plenty of attention in the multimedia community. In person re-identification procedure, choosing a proper distance metric is a crucial aspect [2]. Traditional methods always utilize a uniform learned metric, which ignored specific constraints given by this re-identification task that the learned metric is highly prone to over-fitting [21], and each person holding their unique characteristic brings inconsistency. Therefore, it is obviously inappropriate to merely employ a uniform metric. In this paper, we propose a data-driven metric adaptation method to improve the uniform metric. The key novelty of the approach is that we re-exploits the training data with cross-view consistency to adaptively adjust the metric. Experiments conducted on two standard data sets have validated the effectiveness of the proposed method with a significant improvement over baseline methods.