Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
110 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

CCPA: Long-term Person Re-Identification via Contrastive Clothing and Pose Augmentation (2402.14454v1)

Published 22 Feb 2024 in cs.CV

Abstract: Long-term Person Re-Identification (LRe-ID) aims at matching an individual across cameras after a long period of time, presenting variations in clothing, pose, and viewpoint. In this work, we propose CCPA: Contrastive Clothing and Pose Augmentation framework for LRe-ID. Beyond appearance, CCPA captures body shape information which is cloth-invariant using a Relation Graph Attention Network. Training a robust LRe-ID model requires a wide range of clothing variations and expensive cloth labeling, which is lacked in current LRe-ID datasets. To address this, we perform clothing and pose transfer across identities to generate images of more clothing variations and of different persons wearing similar clothing. The augmented batch of images serve as inputs to our proposed Fine-grained Contrastive Losses, which not only supervise the Re-ID model to learn discriminative person embeddings under long-term scenarios but also ensure in-distribution data generation. Results on LRe-ID datasets demonstrate the effectiveness of our CCPA framework.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (38)
  1. “Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline),” in ECCV, 2018, p. 501–518.
  2. “Omni-scale feature learning for person re-identification,” in ICCV, 2019, pp. 3701–3711.
  3. “Attention-aware compositional network for person re-identification,” in CVPR, 2018, pp. 2119–2128.
  4. “Clothes-changing person re-identification with rgb modality only,” in CVPR, 2022, pp. 1050–1059.
  5. “Dcr-reid: Deep component reconstruction for cloth-changing person re-identification,” IEEE TCSVT, 2023.
  6. “Clothing status awareness for long-term person re-identification,” in ICCV, 2021, pp. 11875–11884.
  7. “Clothing-change feature augmentation for person re-identification,” in CVPR, 2023, pp. 22066–22075.
  8. “Good is bad: Causality inspired cloth-debiasing for cloth-changing person re-identification,” in CVPR, 2023, pp. 1472–1481.
  9. “Fine-grained shape-appearance mutual learning for cloth-changing person re-identification,” in CVPR, 2021, pp. 10508–10517.
  10. “Cloth-changing person re-identification from a single image with gait prediction and regularization,” in CVPR, 2022, pp. 14258–14267.
  11. “Person re-identification by contour sketch under moderate clothing change,” IEEE TPAMI, vol. 43, no. 6, pp. 2029–2046, 2021.
  12. “Deep shape-aware person re-identification for overcoming moderate clothing changes,” IEEE TMM, vol. 24, pp. 4285–4300, 2022.
  13. “Long-term cloth-changing person re-identification,” in ACCV, 2021, pp. 71–88.
  14. “Relational graph attention networks,” arXiv preprint arXiv:1904.05811, 2019.
  15. “Fd-gan: Pose-guided feature distilling gan for robust person re-identification,” NeurIPS, 2018.
  16. “Pose-driven attention-guided image generation for person re-identification,” Pattern Recognition, vol. 137, pp. 109246, 2023.
  17. “Multi-pseudo regularized label for generated data in person re-identification,” IEEE TIP, vol. 28, no. 3, pp. 1391–1403, 2018.
  18. “Unsupervised person re-identification by deep learning tracklet association,” in ECCV, 2018, pp. 772–788.
  19. “Joint generative and contrastive learning for unsupervised person re-identification,” in CVPR, 2021, pp. 2004–2013.
  20. “Unsupervised person re-identification in aerial imagery,” in WACV Workshops, 2024, pp. 260–269.
  21. “Attention-based shape and gait representations learning for video-based cloth-changing person re-identification,” in VISIGRAPP (2: VISAPP), 2024, pp. 80–89.
  22. “A pose-sensitive embedding for person re-identification with expanded cross neighborhood re-ranking,” in CVPR, 2017, pp. 420–429.
  23. “Viewpoint transform matching model for person re-identification,” Neurocomputing, vol. 433, pp. 19–27, 2021.
  24. “Aware loss with angular regularization for person re-identification,” in AAAI, 2020, vol. 34, pp. 13114–13121.
  25. “Rethinking reid: Multi-feature fusion person re-identification based on orientation constraints,” in ICPR, 2021, pp. 1904–1911.
  26. “Learning 3d shape feature for texture-insensitive person re-identification,” in CVPR, 2021, pp. 8142–8151.
  27. “Temporal 3d shape modeling for video-based cloth-changing person re-identification,” in WACV Workshops, 2024, pp. 173–182.
  28. “Generalizing a person retrieval model hetero-and homogeneously,” in ECCV, 2018, pp. 172–188.
  29. “Invariance matters: Exemplar memory for domain adaptive person re-identification,” in CVPR, 2019.
  30. “Realtime multi-person 2d pose estimation using part affinity fields,” in CVPR, 2017, pp. 7291–7299.
  31. “Graph attention networks,” arXiv preprint arXiv:1710.10903, 2017.
  32. “Mebow: Monocular estimation of body orientation in the wild,” in CVPR, 2020.
  33. “Deep residual learning for image recognition,” in CVPR, 2016.
  34. “Imagenet: A large-scale hierarchical image database,” in CVPR, 2009, pp. 248–255.
  35. “Joint discriminative and generative learning for person re-identification,” in CVPR, 2019, pp. 2133–2142.
  36. “Contrastive viewpoint-aware shape learning for long-term person re-identification,” in WACV, 2024, pp. 1041–1049.
  37. “Learning shape representations for person re-identification under clothing change,” in WACV, 2021, pp. 2431–2440.
  38. “Co-attention aligned mutual cross-attention for cloth-changing person re-identification,” in ACCV, 2022, pp. 2270–2288.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (2)
  1. Vuong D. Nguyen (4 papers)
  2. Shishir K. Shah (6 papers)
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now