GLAD: Global-Local-Alignment Descriptor for Pedestrian Retrieval

By Longhui Wei, Shiliang Zhang, Hantao Yao, Wen Gao, Qi Tian. ACM Multimedia, 2017

409 citations recorded by the source corpus.

The huge variance of human pose and the misalignment of detected human images significantly increase the difficulty of person Re-Identification (Re-ID). Moreover, efficient Re-ID systems are required to cope with the massive visual data being produced by video surveillance systems. Targeting to solve these problems, this work proposes a Global-Local-Alignment Descriptor (GLAD) and an efficient indexing and retrieval framework, respectively. GLAD explicitly leverages the local and global cues in human body to generate a discriminative and robust representation. It consists of part extraction and descriptor learning modules, where several part regions are first detected and then deep neural networks are designed for representation learning on both the local and global regions. A hierarchical indexing and retrieval framework is designed to eliminate the huge redundancy in the gallery set, and accelerate the online Re-ID procedure. Extensive experimental results show GLAD achieves competitive accuracy compared to the state-of-the-art methods. Our retrieval framework significantly accelerates the online Re-ID procedure without loss of accuracy. Therefore, this work has potential to work better on person Re-ID tasks in real scenarios.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/bdd3e33cb78965bcd15d69b7902c813b2ac864ff

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

1. Yi Yang
2. Liang Zheng
3. Zhedong Zheng
4. Shiliang Zhang
5. Qi Tian
6. Tao A. Xiang
7. Wen Gao
8. Shengjin Wang

In Computer Vision and Pattern Recognition

1. Yi Yang
2. Dacheng Tao
3. Luc Van Gool
4. Qi Tian
5. Tao A. Xiang
6. Xiaogang Wang
7. Liang Zheng
8. Ming-Hsuan Yang

Citations in the corpus

• Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training (2018, background): “Our method achieves the same result with MGN on [26] 86.9 95.7 - - PCB+RPP [24] 81.6 93.8 97.5 98.5 PCB [24] 77.4 92.3 97.2 98.2 GLAD* [27] 73.9 89.9 - - MultiScale [4] 73.1 88.9 - - PartLoss [29] 69.3 88.2 - - PDC* [22] 63.4 84.4 92.7 94.9 MultiLoss [13] 64.4 83.9 - - PAR [31] 63.4 81.0 92.0 94.7…”
• ABD-Net: Attentive but Diverse Person Re-Identification (2019, positive): “Many part-based methods have achieved superior performance [2, 3, 26, 27, 28, 4, 5, 29, 6, 30, 31, 8, 32].”
• Deep Representation Learning With Part Loss for Person Re-Identification (2017, background): “Therefore, many works are proposed to fuse body part representations for person ReID [27], [41]–[44].”
• CityFlow: A City-Scale Benchmark for Multi-Target Multi-Camera Vehicle Tracking and Re-Identification (2019, neutral): “ng the overall best for both mAP and rank-1. Method Market1501 DukeMTMC-reID MSMT17 [59] [35,63] [47] HA-CNN [24] 75.6 (90.9) 63.2 (80.1) 37.2 (64.7) MLFN [7] 74.3 (90.1) 63.2 (81.1) 37.2 (66.4) GLAD [50] - - 34.0 (61.4) Res50+Xent 75.3 (90.8) 64.0 (81.0) 38.4 (69.6) Res50M+Xent 76.0 (90.2) 64.0 (81”
• Joint Discriminative and Generative Learning for Person Re-Identification (2019, methodology): “To further reduce the influence from intra-class variations, a number of existing methods adopt part-based matching or ensemble to explicitly align and compensate the variations [35,37,47,52,57].”
• Perceive Where to Focus: Learning Visibility-Aware Part-Level Features for Partial Person Re-Identification (2019, background): “Recent works [23, 12, 26, 29, 22, 28, 16] further advance the state of the art on holistic person re-ID, through learning part-level deep features.”
• AANet: Attribute Attention Network for Person Re-Identifications (2019, methodology): “[22] also adopted the human pose estimation, or key point detection approach, in his Global-Local-Alignment Descriptor (GLAD) algorithm.”
• TransReID: Transformer-based Object Re-Identification (2021, methodology): “Using parsing or keypoint estimation to align different parts or two objects has also been proven effective for both person and vehicle ReID [22, 25, 39, 26].”
• Robust Person Re-Identification by Modelling Feature Uncertainty (2019, neutral): “To overcome misalignment and pose variations, several works use pose detectors to identify body parts on which part-specific features are learned [34, 25, 31, 32].”
• Pose-Normalized Image Generation for Person Re-identification (2017, background): “A number of models [39, 62, 58, 57, 23, 50, 45] are proposed to address this problem.”
• Omni-Scale Feature Learning for Person Re-Identification (2019, background): “scale/multi-stream architecture, namely those with a Inception or ResNeXt backbone [27, 44, 4, 60, 2, 42], OSNet is clearly superior.”
• Bag of Tricks and a Strong Baseline for Deep Person Re-Identification (2019, methodology): “ls is evaluated on cross-domain datasets. M!D means that we train the model on Market1501 and evaluate it on DukeMTMC-reID. 4326 Market1501 DukeMTMC Type Method N f r = 1 mAP r = 1 mAP Poseguided GLAD[19] 4 89.9 73.9 - - PIE [23] 3 87.7 69.0 79.8 62.0 PSE [13] 3 78.7 56.0 - - Maskguided SPReID [7] 5”
• Person Transfer GAN to Bridge Domain Gap for Person Re-identification (2017, methodology): “However, the best performance achieved by GLAD, e.g. , mAP of 34%, is still substantially lower than its reported performance on other datasets, e.g. , 73.9% on Market .”
• Bilinear Attention Networks for Person Retrieval (2019, neutral): “4 Glad* [47] 34.”