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

Multi-Memory Matching for Unsupervised Visible-Infrared Person Re-Identification (2401.06825v2)

Published 12 Jan 2024 in cs.CV

Abstract: Unsupervised visible-infrared person re-identification (USL-VI-ReID) is a promising yet challenging retrieval task. The key challenges in USL-VI-ReID are to effectively generate pseudo-labels and establish pseudo-label correspondences across modalities without relying on any prior annotations. Recently, clustered pseudo-label methods have gained more attention in USL-VI-ReID. However, previous methods fell short of fully exploiting the individual nuances, as they simply utilized a single memory that represented an identity to establish cross-modality correspondences, resulting in ambiguous cross-modality correspondences. To address the problem, we propose a Multi-Memory Matching (MMM) framework for USL-VI-ReID. We first design a Cross-Modality Clustering (CMC) module to generate the pseudo-labels through clustering together both two modality samples. To associate cross-modality clustered pseudo-labels, we design a Multi-Memory Learning and Matching (MMLM) module, ensuring that optimization explicitly focuses on the nuances of individual perspectives and establishes reliable cross-modality correspondences. Finally, we design a Soft Cluster-level Alignment (SCA) module to narrow the modality gap while mitigating the effect of noise pseudo-labels through a soft many-to-many alignment strategy. Extensive experiments on the public SYSU-MM01 and RegDB datasets demonstrate the reliability of the established cross-modality correspondences and the effectiveness of our MMM. The source codes will be released.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (58)
  1. Visible-infrared person re-identification using privileged intermediate information. In ECCV, pages 720–737, 2022.
  2. A closer look at memorization in deep networks. In ICML, pages 233–242, 2017.
  3. ICE: inter-instance contrastive encoding for unsupervised person re-identification. In ICCV, pages 14940–14949, 2021a.
  4. Neural feature search for rgb-infrared person re-identification. In CVPR, pages 587–597, 2021b.
  5. Unveiling the power of clip in unsupervised visible-infrared person re-identification. In ACM MM, pages 3667–3675, 2023.
  6. Unsupervised visible-infrared person reid by collaborative learning with neighbor-guided label refinement. ArXiv:2305.12711, 2023.
  7. Part-based pseudo label refinement for unsupervised person re-identification. In CVPR, pages 7298–7308, 2022.
  8. Hi-cmd: Hierarchical cross-modality disentanglement for visible-infrared person re-identification. In CVPR, pages 10254–10263, 2020.
  9. Cluster contrast for unsupervised person re-identification. In ACCV, pages 319–337, 2022.
  10. A density-based algorithm for discovering clusters in large spatial databases with noise. In KDD, pages 226–231, 1996.
  11. Shape-erased feature learning for visible-infrared person re-identification. In CVPR, pages 22752–22761, 2023.
  12. Self-similarity grouping: A simple unsupervised cross domain adaptation approach for person re-identification. In ICCV, pages 6111–6120, 2019.
  13. Mutual mean-teaching: Pseudo label refinery for unsupervised domain adaptation on person re-identification. In ICLR, 2020a.
  14. Self-paced contrastive learning with hybrid memory for domain adaptive object re-id. In NeurIPS, 2020b.
  15. A kernel two-sample test. The Journal of Machine Learning Research, 13(1):723–773, 2012.
  16. Efficient bilateral cross-modality cluster matching for unsupervised visible-infrared person reid. ArXiv:2305.12673, 2023.
  17. Modality-adaptive mixup and invariant decomposition for rgb-infrared person re-identification. In AAAI, pages 1034–1042, 2022.
  18. Comparing partitions. Journal of classification, 2:193–218, 1985.
  19. Partmix: Regularization strategy to learn part discovery for visible-infrared person re-identification. In CVPR, pages 18621–18632, 2023.
  20. Infrared-visible cross-modal person re-identification with an X modality. In AAAI, pages 4610–4617, 2020.
  21. Homogeneous-to-heterogeneous: Unsupervised learning for rgb-infrared person re-identification. IEEE Trans. Image Process., 30:6392–6407, 2021.
  22. Multi-task mid-level feature alignment network for unsupervised cross-dataset person re-identification. In BMVC, page 9, 2018.
  23. A bottom-up clustering approach to unsupervised person re-identification. In AAAI, pages 8738–8745, 2019.
  24. Unsupervised person re-identification via softened similarity learning. In CVPR, pages 3387–3396, 2020.
  25. Learning memory-augmented unidirectional metrics for cross-modality person re-identification. In CVPR, pages 19344–19353, 2022.
  26. James MacQueen et al. Some methods for classification and analysis of multivariate observations. In Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, pages 281–297, 1967.
  27. Person recognition system based on a combination of body images from visible light and thermal cameras. Sensors, 17(3):605, 2017.
  28. Learning by aligning: Visible-infrared person re-identification using cross-modal correspondences. In ICCV, pages 12026–12035, 2021.
  29. Learning transferable visual models from natural language supervision. In ICML, pages 8748–8763, 2021.
  30. Dual pseudo-labels interactive self-training for semi-supervised visible-infrared person re-identification. In ICCV, pages 11218–11228, 2023.
  31. Unsupervised domain adaptive re-identification: Theory and practice. Pattern Recognit., 102:107173, 2020.
  32. Not all pixels are matched: Dense contrastive learning for cross-modality person re-identification. In ACM MM, pages 5333–5341, 2022.
  33. Unsupervised person re-identification via multi-label classification. In CVPR, pages 10978–10987, 2020.
  34. Rgb-infrared cross-modality person re-identification via joint pixel and feature alignment. In ICCV, pages 3622–3631, 2019.
  35. Cross-modality paired-images generation and augmentation for rgb-infrared person re-identification. Neural Networks, 128:294–304, 2020a.
  36. Cross-modality paired-images generation and augmentation for rgb-infrared person re-identification. Neural Networks, 128:294–304, 2020b.
  37. Transferable joint attribute-identity deep learning for unsupervised person re-identification. In CVPR, pages 2275–2284, 2018.
  38. Optimal transport for label-efficient visible-infrared person re-identification. In ECCV, pages 93–109, 2022.
  39. Syncretic modality collaborative learning for visible infrared person re-identification. In ICCV, pages 225–234, 2021.
  40. Rgb-infrared cross-modality person re-identification. In ICCV, pages 5390–5399, 2017.
  41. Discover cross-modality nuances for visible-infrared person re-identification. In CVPR, pages 4330–4339, 2021.
  42. Unsupervised visible-infrared person re-identification via progressive graph matching and alternate learning. In CVPR, pages 9548–9558, 2023.
  43. Intra-inter camera similarity for unsupervised person re-identification. In CVPR, pages 11926–11935, 2021.
  44. Augmented dual-contrastive aggregation learning for unsupervised visible-infrared person re-identification. In ACM MM, pages 2843–2851, 2022a.
  45. Dual consistency-constrained learning for unsupervised visible-infrared person re-identification. IEEE Transactions on Information Forensics and Security, 2023a.
  46. Translation, association and augmentation: Learning cross-modality re-identification from single-modality annotation. IEEE Transactions on Image Processing, 32:5099–5113, 2023b.
  47. Towards grand unified representation learning for unsupervised visible-infrared person re-identification. In ICCV, pages 11069–11079, 2023c.
  48. Learning with twin noisy labels for visible-infrared person re-identification. In CVPR, pages 14288–14297, 2022b.
  49. Visible thermal person re-identification via dual-constrained top-ranking. In IJCAI, pages 1092–1099, 2018a.
  50. Visible thermal person re-identification via dual-constrained top-ranking. In IJCAI, pages 1092–1099, 2018b.
  51. Dynamic dual-attentive aggregation learning for visible-infrared person re-identification. In ECCV, pages 229–247, 2020.
  52. Channel augmented joint learning for visible-infrared recognition. In ICCV, pages 13547–13556, 2021.
  53. Deep learning for person re-identification: A survey and outlook. IEEE Trans. Pattern Anal. Mach. Intell., pages 2872–2893, 2022.
  54. Multiple expert brainstorming for domain adaptive person re-identification. In ECCV, pages 594–611, 2020.
  55. Camera contrast learning for unsupervised person re-identification. IEEE Trans. Circuits Syst. Video Technol., 33(8):4096–4107, 2023.
  56. Fmcnet: Feature-level modality compensation for visible-infrared person re-identification. In CVPR, pages 7339–7348, 2022.
  57. Diverse embedding expansion network and low-light cross-modality benchmark for visible-infrared person re-identification. In CVPR, pages 2153–2162, 2023.
  58. Towards a unified middle modality learning for visible-infrared person re-identification. In ACM MM, pages 788–796, 2021.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (7)
  1. Jiangming Shi (4 papers)
  2. Xiangbo Yin (4 papers)
  3. Yeyun Chen (1 paper)
  4. Yachao Zhang (21 papers)
  5. Zhizhong Zhang (42 papers)
  6. Yuan Xie (188 papers)
  7. Yanyun Qu (39 papers)
Citations (11)
View on Semantic Scholar

Summary

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

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