Hierarchical Bi-Directional Feature Perception Network for Person Re-Identification

Liu, Zhipu; Zhang, Lei; Yang, Yang

doi:10.1145/3394171.3413689

Cited by 14 publications

(4 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Interaction In Transformersmentioning

confidence: 99%

“…Furthermore, for image-based person Re-ID, previous works [4,26,47,59] explore the effectiveness of hierarchical features of CNNs. For example, some works [4,26] utilize attention-based structures to adaptively fuse multi-level features. Zhou et al [59] propose an unified aggregation gate to dynamically fuse multi-scale features with different channel-wise weights.…”

Section: Interaction In Transformersmentioning

confidence: 99%

“…2(b). Liu et al [26] utilize correlation maps of cross-level features to enrich high-level features and learn salient and specific information in low-level features. However, simple aggregation structures will lead to poor performance [47] for Re-ID.…”

Section: Deep Feature Aggregationmentioning

confidence: 99%

See 2 more Smart Citations

HAT: Hierarchical Aggregation Transformers for Person Re-identification

Zhang

et al. 2021

Preprint

View full text Add to dashboard Cite

Recently, with the advance of deep Convolutional Neural Networks (CNNs), person Re-Identification (Re-ID) has witnessed great success in various applications. However, with limited receptive fields of CNNs, it is still challenging to extract discriminative representations in a global view for persons under non-overlapped cameras. Meanwhile, Transformers demonstrate strong abilities of modeling long-range dependencies for spatial and sequential data. In this work, we take advantages of both CNNs and Transformers, and propose a novel learning framework named Hierarchical Aggregation Transformer (HAT) for image-based person Re-ID with high performance.To achieve this goal, we first propose a Deeply Supervised Aggregation (DSA) to recurrently aggregate hierarchical features from CNN backbones. With multi-granularity supervisions, the DSA can enhance multi-scale features for person retrieval, which is very different from previous methods. Then, we introduce a Transformer-based Feature Calibration (TFC) to integrate low-level detail information as the global prior for high-level semantic information. The proposed TFC is inserted to each level of hierarchical features, resulting in great performance improvements. To our best knowledge, this work is the first to take advantages of both CNNs and Transformers for image-based person Re-ID. Comprehensive experiments on four large-scale Re-ID benchmarks demonstrate that our method shows better results than several state-of-the-art methods. The code is released at https://github.com/AI-Zhpp/HAT.

show abstract

Section: Interaction In Transformersmentioning

confidence: 99%

Section: Interaction In Transformersmentioning

confidence: 99%

See 1 more Smart Citation

HAT: Hierarchical Aggregation Transformers for Person Re-identification

Zhang

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…the other state-of-the-art methods, the proposed LDS achieves competitive results. There are other three multi-branch methods, PTL [63], CAMA [62], and HBFP-Net [33]. These methods employed the feature map from different layers to form a rich feature representation.…”

Section: Performances On Market1501 In Table 1 Compared Tomentioning

confidence: 99%

Learning to Disentangle Scenes for Person Re-identification

Zang,

Li,

Gao

et al. 2021

Preprint

View full text Add to dashboard Cite

There are many challenging problems in the person re-identification (ReID) task, such as the occlusion and scale variation. Existing works usually tried to solve them by employing a one-branch network. This one-branch network needs to be robust to various challenging problems, which makes this network overburdened. This paper proposes to divide-and-conquer the ReID task. For this purpose, we employ several self-supervision operations to simulate different challenging problems and handle each challenging problem using different networks. Concretely, we use the random erasing operation and propose a novel random scaling operation to generate new images with controllable characteristics. A general multi-branch network, including one master branch and two servant branches, is introduced to handle different scenes. These branches learn collaboratively and achieve different perceptive abilities. In this way, the complex scenes in the ReID task are effectively disentangled, and the burden of each branch is relieved. The results from extensive experiments demonstrate that the proposed method achieves state-of-the-art performances on three ReID benchmarks and two occluded ReID benchmarks. Ablation study also shows that the proposed scheme and operations significantly improve the performance in various scenes.

show abstract

Multi-scale Context Aggregation for Video-Based Person Re-Identification

Wu,

Zhang,

et al. 2023

Communications in Computer and Information Science

View full text Add to dashboard Cite

Hierarchical Bi-Directional Feature Perception Network for Person Re-Identification

Cited by 14 publications

References 42 publications

HAT: Hierarchical Aggregation Transformers for Person Re-identification

HAT: Hierarchical Aggregation Transformers for Person Re-identification

Learning to Disentangle Scenes for Person Re-identification

Multi-scale Context Aggregation for Video-Based Person Re-Identification

Contact Info

Product

Resources

About