71 Hits in 4.2 sec


Longhui Wei, Shiliang Zhang, Hantao Yao, Wen Gao, Qi Tian
2017 Proceedings of the 2017 ACM on Multimedia Conference - MM '17  
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.  ...  GLOBAL-LOCAL-ALIGNMENT DESCRIPTOR The framework of GLAD extraction is summarized in Fig. 2 .  ... 
doi:10.1145/3123266.3123279 dblp:conf/mm/WeiZY0T17 fatcat:z63h634unngtpebaxuxmvy3y4y

Joint Learning of Body and Part Representation for Person Re-Identification

Yuanyuan Wang, Zhijian Wang, Wenjing Jia, Xiangjian He, Mingxin Jiang
2018 IEEE Access  
features from pedestrian images.  ...  In this paper, following the recent success of part-based models, in order to generate a discriminative and robust feature representation, we first propose to learn global and weighted local body-part  ...  When learning local body-part features, Wei et al. [2] proposed the framework of global-local-alignment descriptor (GLAD), which cropped body parts before extracting local features.  ... 
doi:10.1109/access.2018.2864588 fatcat:vmsd3bjfwbdmdemv3ovjv5amke

Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline) [article]

Yifan Sun, Liang Zheng, Yi Yang, Qi Tian, Shengjin Wang
2018 arXiv   pre-print
Employing part-level features for pedestrian image description offers fine-grained information and has been verified as beneficial for person retrieval in very recent literature.  ...  Given an image input, it outputs a convolutional descriptor consisting of several part-level features.  ...  In contrast to the proposed PCB, the IDE model learns a global descriptor.  ... 
arXiv:1711.09349v3 fatcat:wmcahe2w5zdnvcambuhuskidyq

Beyond Part Models: Person Retrieval with Refined Part Pooling (and A Strong Convolutional Baseline) [chapter]

Yifan Sun, Liang Zheng, Yi Yang, Qi Tian, Shengjin Wang
2018 Lecture Notes in Computer Science  
Employing part-level features offers fine-grained information for pedestrian image description.A prerequisite of part discovery is that each part should be well located.  ...  With a uniform partition strategy, PCB achieves competitive results with the state-of-the-art methods, proving itself as a strong convolutional baseline for person retrieval.  ...  In contrast to the proposed PCB, the IDE model learns a global descriptor.  ... 
doi:10.1007/978-3-030-01225-0_30 fatcat:xkhwvczmgrfmfgkxbw4zifquue

Learning Multi-scale Features and Batch-normalized Global Features for Person Re-identification

Zongjing Cao, Hyo Jong Lee
2020 IEEE Access  
His more recent research focuses on image retrieval and person reidentification.  ...  In the inference phase, we choose the batch-normalized global feature as the pedestrian descriptor that is used to perform the person re-ID task.  ...  Before deep learning methods dominated the re-ID research community, many hand-crafted algorithms were developed to learn the global and local features of pedestrians.  ... 
doi:10.1109/access.2020.3029594 fatcat:6opjvzmy6bgujewgituqkgv7tm

Person re-identification with fusion of hand-crafted and deep pose-based body region features [article]

Jubin Johnson, Shunsuke Yasugi, Yoichi Sugino, Sugiri Pranata and Shengmei Shen
2018 arXiv   pre-print
However, identifying unseen persons is critical for a good re-ID algorithm. Moreover, the misalignment be- tween person crops to detection errors or pose variations leads to poor feature matching.  ...  In this work, we present a fusion of handcrafted features and deep feature representa- tion learned using multiple body parts to complement the global body features that achieves high performance on un  ...  A Global-Local Alignment Descriptor (GLAD) [37] does not directly align pedestrians, but rather detects key pose points and extracts local features from corresponding regions.  ... 
arXiv:1803.10630v1 fatcat:3fnnkry7qnamlng66kxpsg5jiy

AANet: Attribute Attention Network for Person Re-Identifications [article]

Chiat-Pin Tay, Sharmili Roy, Kim-Hui Yap
2019 arXiv   pre-print
The AANet consists of a global person ID task, a part detection task and a crucial attribute detection task.  ...  In addition, AANet can perform person attribute prediction (e.g., gender, hair length, clothing length etc.), and localize the attributes in the query image.  ...  [24] also adopted the human pose estimation, or key point detection approach, in his Global-Local-Alignment Descriptor (GLAD) algorithm.  ... 
arXiv:1912.09021v1 fatcat:ed3wsvjplbgptej4re32ro2n5e

AlignedReID: Surpassing Human-Level Performance in Person Re-Identification [article]

Xuan Zhang, Hao Luo, Xing Fan, Weilai Xiang, Yixiao Sun, Qiqi Xiao, Wei Jiang, Chi Zhang, Jian Sun
2018 arXiv   pre-print
Global feature learning benefits greatly from local feature learning, which performs an alignment/matching by calculating the shortest path between two sets of local features, without requiring extra supervision  ...  In this paper, we propose a novel method called AlignedReID that extracts a global feature which is jointly learned with local features.  ...  A Global-Local-Alignment Descriptor (GLAD) [37] does not directly align pedestrians, but rather detects key pose points and extracts local features from corresponding regions.  ... 
arXiv:1711.08184v2 fatcat:d22zbqrbrvhcdnbaa5w47ylza4

CA3Net: Contextual-Attentional Attribute-Appearance Network for Person Re-Identification [article]

Jiawei Liu and Zheng-Jun Zha and Hongtao Xie and Zhiwei Xiong and Yongdong Zhang
2018 arXiv   pre-print
Deep learning techniques have been applied for person re-identification recently, towards learning representation of pedestrian appearance.  ...  This paper presents a novel Contextual-Attentional Attribute-Appearance Network (CA3Net) for person re-identification.  ...  Conventional person re-identification approaches are mainly based on hand-crafted descriptors of pedestrian appearance [6, 34] , such as Symmetry-Driven Accumulation of Local Features (SDALF) [7] , Local  ... 
arXiv:1811.07544v1 fatcat:qsgoqomymjbuzmqe3wk3nljarm

Person Re-identification based on Two-Stream Network with Attention and Pose Features

XiaoWei Gong, SuoGuo Zhu
2019 IEEE Access  
For more information, see VOLUME 7, 2019  ...  As a consequence, we fusion the mid-level and high-level features, and then correlate global information through self-attention.  ...  [3] proposed a global-local-alignment Descriptor to solve the problem of pedestrian posture change.  ... 
doi:10.1109/access.2019.2935116 fatcat:xx2nbv7yhrghzp2ly6me6pjjcm

Deep learning-based person re-identification methods: A survey and outlook of recent works [article]

Zhangqiang Ming, Min Zhu, Xiangkun Wang, Jiamin Zhu, Junlong Cheng, Chengrui Gao, Yong Yang, Xiaoyong Wei
2021 arXiv   pre-print
The main research goal of person Re-ID is to retrieve persons with the same identity from different cameras.  ...  deep metric learning, local feature learning, generative adversarial learning and sequence feature learning.  ...  [84] design a global-local dynamic feature alignment network (GLDFA-Net) framework, which contains both global and local branches, their network structure is shown in Figure 7 .  ... 
arXiv:2110.04764v4 fatcat:4pdg4zwr4fbvtjhxwnxauytkwq

Do Not Disturb Me: Person Re-identification Under the Interference of Other Pedestrians [article]

Shizhen Zhao, Changxin Gao, Jun Zhang, Hao Cheng, Chuchu Han, Xinyang Jiang, Xiaowei Guo, Wei-Shi Zheng, Nong Sang, Xing Sun
2020 arXiv   pre-print
This will lead to wrong retrieval results. To address this problem, this paper presents a novel deep network termed Pedestrian-Interference Suppression Network (PISNet).  ...  In the conventional person Re-ID setting, it is widely assumed that cropped person images are for each individual.  ...  Wei, L., Zhang, S., Yao, H., Gao, W., Tian, Q.: Glad: global-local-alignment descriptor for pedestrian retrieval. In: ACM on Multimedia Conference (ACM MM) (2017) 35.  ... 
arXiv:2008.06963v1 fatcat:ri2ti5xpi5datgewihqxu43l6u

Learning Deep Representations by Mutual Information for Person Re-identification [article]

Peng Chen, Tong Jia, Pengfei Wu, Jianjun Wu, Dongyue Chen
2019 arXiv   pre-print
To evaluate the effectiveness of the DIM network, we propose end-to-end Global-DIM and Local-DIM models.  ...  Additionally, the DIM network provides a new solution for cross-dataset unsupervised ReID issue as it needs no extra labels.  ...  [9] proposed a Global-Local-Alignment Descriptor (GLAD) to partition the pedestrian. Su et al. [10] utilized pedestrian part cues to boost feature representations.  ... 
arXiv:1908.05860v1 fatcat:cl2gipfgwzbs3en3i6lk7fzoga

Improving semantic part features for person re-identification with supervised non-local similarity

Yifan Sun, Zhaopeng Dou, Yali Li, Shengjin Wang
2020 Tsinghua Science and Technology  
To facilitate part alignment, which is a prerequisite for learning part-level features, a popular approach is to detect semantic parts with the use of human parsing or pose estimation.  ...  In response, this paper proposes a novel part feature learning method for re-ID, that suppresses the impact of noisy semantic part detection through Supervised Non-local Similarity (SNS) learning.  ...  For example, GLAD [2] uses Open Pose [16] to detect several landmarks and joints on a pedestrian and then extract the head, upper-body, and lower-body with bounding boxes.  ... 
doi:10.26599/tst.2019.9010024 fatcat:zj73hs2p7rfhhngitac3svrbi4

Receptive Multi-granularity Representation for Person Re-Identification [article]

Guanshuo Wang, Yufeng Yuan, Jiwei Li, Shiming Ge, Xi Zhou
2020 arXiv   pre-print
A key for person re-identification is achieving consistent local details for discriminative representation across variable environments.  ...  Toward this end, the local partitions are adaptively pooled by using significance-balanced activations for uniform stripes.  ...  Unsupervised person re-ID methods are significant for better-generalized pedestrian retrieval. Peng et al.  ... 
arXiv:2008.13450v1 fatcat:bi7lgqfp4vfnrn5aiz4kgmiumi
« Previous Showing results 1 — 15 out of 71 results