3D Skeleton and Two Streams Approach to Person Re-identification Using Optimized Region Matching

Abstract

Person re-identification (Re-ID) is a challenging and arduous task due to non-overlapping views, complex background, and uncontrollable occlusion in video surveillance. An existing method for capturing pedestrian local region information is to divide person regions into horizontal stripes, which may lead to invalid features and erroneous learning. To solve this problem, this paper proposes a 3D skeleton and a two-stream approach to person Re-ID. The first stream of the method uses the 3D skeleton for background filtering and region segmentation. The second stream uses Siamese net to extract the global descriptor. The features of the two streams are fused to preserve the integrity of the person. An optimized region matching method for metric learning is designed. Extensive comparing experiments were conducted with state-of-the-art Re-ID methods on the Market-1501, CUHK03, and DukeMTMC-reID datasets. Experimental results show that the proposed method outperforms the existing methods in recognition accuracy.

Unsupervised Vehicle Re-Identification Based on Cross-Style Semi-Supervised Pre-Training and Feature Cross-Division

“…During the optimization stage, this paper uses the commonly used cross-entropy loss function to optimize the network. Therefore, the total loss function of the framework is shown in Equation (7). Driven by the total loss function, the model in this paper performs outstandingly in addressing the domain gap and reducing the noise carried by images in different datasets.…”

“…These methods typically involve pre-training the model using source domain data and subsequently fine-tuning the pre-trained model on the target domain. Despite the advancements brought about by unsupervised domain adaptation, the performance of vehicle Re-ID still falls short of that achieved by supervised learning methods [7][8][9]. This performance gap can be attributed to the existing domain gap [10] between the source and target domains, as well as the reliance on global features for pseudo-label assignment during fine-tuning.…”

Joint training with local soft attention and dual cross-neighbor label smoothing for unsupervised person re-identification

“…Person re-identification (Re-ID) aims to match queried person images of the same identity across non-overlapping camera views. Person and vehicle Re-ID play vital roles in areas, such as intelligent video surveillance, security, and transport [1][2][3]. In recent years, unsupervised person Re-ID has attracted considerable attention because it does not require massive amounts of manually labeled data.…”

“…Some existing approaches [15][16][17] optimize clustering results by combining global and local training strategies. These methods, despite their effectiveness, ignore two crucial factors in this process: (1) The inadequate capture of local fine-grained features interferes with the clustering results which depend on the metric ranking of the global features extracted by the model. Some approaches ignore the capture of finegrained features in local regions, thus affecting the ability of the entire model to learn discriminant feature embedding from the data such data persons with similar appearances and different identities can be assigned the same label after clustering, which directly affects the training and final results of the model.…”