Pith. sign in

REVIEW 2 major objections 6 minor 2 cited by

Exploring the Camera Bias of Person Re-identification

T0 review · 2 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper claims that per-camera mean and standard deviation normalization of embedding features is a general test-time debiasing step for person re-identification, and that debiased pseudo labels plus discarding single-camera clusters…

desk verdict Solid, well-executed empirical study: camera-specific normalization works on seen/unseen domains when camera labels and enough samples per camera exist, but the abstract overclaims and the paper's own Figure 5 shows the method degrades sharply with sparse cameras. read the letter →

arxiv 2502.10195 v1 pith:QK355SET submitted 2025-02-14 cs.CV cs.AIcs.LG

Myungseo Song , Jin-Woo Park , Jong-Seok Lee This is my paper · ORCID
classification cs.CVcs.AIcs.LG
keywords camerabiaspersonre-identificationfeaturenormalizationunsuperviseddomainshiftpseudolabelstest-timepostprocessingcamera-specificstatistics
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper tries to establish that camera bias in person re-identification is not only a training-time problem: it grows when a model is evaluated on a new camera domain, and a test-time operation on the embedding vectors can remove much of it. The operation replaces each embedding by the per-camera centered and scaled version $\hat{f} = (f - m_c)/\sigma_c$, where $m_c$ and $\sigma_c$ are that camera's mean and standard deviation over the target set. The paper explains the mechanism by showing that only a small number of embedding dimensions respond consistently to camera changes, so subtracting the camera mean in those dimensions is what debiases the representation. It also claims the same idea applies to unsupervised training, where camera-biased pseudo labels poison the model; debiasing the pseudo labels and discarding single-camera clusters yields large accuracy gains, including a 19.3 mAP increase on MSMT17. A sympathetic reader would care because it offers a training-free fix for deployed systems and a minimal modification for unsupervised ones.

What carries the argument

The central object is camera-specific feature normalization (Equation 2), defined as subtracting the per-camera mean embedding and dividing by the per-camera standard deviation. The analysis identifies the load-bearing mechanism: the per-dimension variance of camera mean features is highly uneven, and feature displacement vectors caused by camera changes are consistent across identities in the high-variance dimensions, so mean-centering those dimensions accounts for most of the improvement while rotation-based whitening does not. The same mechanism is then reused as group-specific normalization for low-level image properties and body angles, and as a debiasing step inside unsupervised pseudo-label generation.

What would settle it

Run the normalization on the same benchmark but with camera labels randomly permuted before computing statistics; if accuracy still rises as much as with true labels, the camera-specific mechanism is not what carries the gain. Equivalently, reproducing the paper's sample-size curve on Market-1501 with a model trained on MSMT17 should show mAP near 7.6 with five samples per camera and around 62 with 100 or more; a result that does not show that cliff would falsify the stated practical boundary.

Watch

Extended reading notes

Core claim

Stated on the paper's own terms, the central discovery is that camera-specific feature normalization is a general debiasing method for ReID embeddings on unseen domains: across every model tested, whether supervised, unsupervised, camera-aware, camera-agnostic, CNN, or transformer, the normalization improves retrieval accuracy and reduces the NMI between cluster labels and camera labels. The reason is dimensional: camera changes move features consistently along a small set of camera-sensitive dimensions, so centering those dimensions dominates the gain, and scaling adds a smaller extra benefit. The same normalization extends to finer bias factors such as low-level image properties and body angle, where jointly conditioning on camera and the factor outperforms camera alone. A second claim is that unsupervised models inherit a large camera bias from their pseudo labels; generating pseudo labels from normalized features and removing single-camera clusters improves existing unsupervised algorithms substantially, for example raising CC on MSMT17 from 29.8 to 49.1 mAP.

Load-bearing premise

The load-bearing premise is that the deployment knows which camera took each target image and has enough images per camera, roughly 25 to 100, to estimate reliable per-camera means and standard deviations; with only five per camera the reported gains collapse to near baseline, 7.6 mAP in the paper's Figure 5.

Editorial extensions

If this is right

  • On unseen target domains, the normalization improves mAP and Rank-1 for all evaluated ReID models, with examples such as TransReID-SSL on Market-1501 going from 53.6 to 62.3 mAP.
  • Debiasing is carried mostly by a small set of camera-sensitive dimensions; the standard-deviation scaling adds a modest extra gain, and ZCA whitening does not match it.
  • Combining camera labels with property or body-angle groups in the normalization can outperform camera-only normalization, e.g., about 1.5 mAP gain for area-grouped normalization on CUHK03-NP.
  • In unsupervised training, debiased pseudo labels and discarding single-camera clusters improve existing algorithms, e.g., CC on MSMT17 from 29.8 to 49.1 mAP, with the largest gain coming from both strategies together.
  • The normalization is compatible with existing postprocessing methods such as DBA, AQE, and reranking, and improves all their metrics rather than trading off early ranks.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: because gains saturate near 100 samples per camera, the normalization doubles as a site-calibration tool; a short survey of a new camera network may give most of the benefit without retraining.
  • Editorial inference: the same cluster-discarding logic could transfer to other clustering-based representation learning tasks where cluster purity is corrupted by a dominant nuisance variable, not just camera.
  • Editorial inference: the paper's static-statistics assumption suggests an online or adaptive variant with a running per-camera mean and standard deviation would be the natural next test for deployments where camera population or lighting shifts over time.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. The paper studies camera bias in person re-identification (ReID) models, focusing on unseen target domains. It measures bias via NMI between cluster labels and camera labels across supervised, unsupervised, camera-aware, and domain-generalizable models. It revisits camera-specific feature normalization (subtracting per-camera mean and dividing by per-camera standard deviation, Eq. 2) as a postprocessing step, analyzes the feature-space mechanism (dimension-dependent sensitivity and consistent displacement vectors), extends the normalization to low-level image properties and body angle, and evaluates it on a wide range of models and benchmarks. In a second part, it argues that unsupervised ReID is harmed by camera-biased pseudo labels and single-camera clusters, and proposes two training modifications: clustering on normalized features and discarding single-camera clusters. Experiments show large gains, e.g., 19.3 mAP for CC on MSMT17.

Significance. The empirical scope is a strength: 13 models, multiple backbones, four person benchmarks plus a vehicle dataset, and ablations isolating centering, scaling, sample count, clustering epsilon, and combination with DBA/AQE/reranking. The paper honestly reports where normalization does not help (seen domains for supervised and camera-aware models) and provides a sample-count curve (Fig. 5) that bounds the method's regime. The toy experiments in Sec. 5.2 give a clean demonstration that camera-biased pseudo labels at equal accuracy hurt training. Source code is promised in the supplementary material. If the claims hold, the work provides a simple, broadly applicable debiasing tool and practical fixes for unsupervised ReID; the main caveat is that the tool's applicability depends on camera metadata and sufficient per-camera sample volume.

major comments (2)
  1. [Abstract, §4.4, Sec. J, Fig. 5] The blanket claim that normalization "significantly improves the performance of all models on the unseen domain" is not scoped to the data-availability condition under which Eq. 2 is usable. Fig. 5 shows that with 5 samples per camera normalized mAP is 7.6 versus 53.6 without normalization, at 10 samples it is 37.0, and only at 25 samples does it roughly tie the baseline; meaningful gains require roughly 50-100 samples per camera. Section J lists only compute cost as a limitation, not the requirement of known camera labels and sufficient per-camera samples. Please add this condition to the limitations and qualify the abstract's "test-time postprocessing" characterization, since in a streaming deployment the query itself is used to estimate its camera's statistics and the method can lose its benefit or even hurt performance.
  2. [Table 3, Table 6, §5.4] The reproduced baselines marked with "*" differ from the official numbers reported in the same tables (e.g., CC baseline on MSMT17 is 29.8 for CC* vs 33.0 for CC, and PPLR* is 27.2 vs PPLR at 31.4). While comparisons within the reproduced runs are internally consistent, the headline gain "19.3% mAP increase for CC" is computed against a reproduced baseline. Please report multiple-seed variance or explicitly compare against the official baseline numbers, so readers can assess the sensitivity of the claimed gains to the reproduced setup.
minor comments (6)
  1. [Algorithm 1] The algorithm title contains a typo: "modificaitons" should be "modifications".
  2. [Sec. 3, Table 1] The sentence "Relatively, the recent supervised models exhibit less debiased results on the training domain" appears to mean "more debiased"; please rephrase for clarity.
  3. [Fig. 5] The caption and axis label should clarify what "using all samples" means in the random sampling protocol, since the total number of samples per camera in Market-1501 is not 1500.
  4. [Eq. 2] Consider adding a small epsilon to the denominator in Eq. 2 to avoid numerical instability when some feature dimensions have near-zero variance, or state explicitly that no epsilon was needed in the reported experiments.
  5. [Table 3] The color coding (white/gray/red backgrounds) for seen versus unseen domains is difficult to read in grayscale; adding explicit symbols or a separate column for seen/unseen would make the table self-contained.
  6. [§4.2] The analysis of camera-sensitive dimensions is descriptive and uses ground-truth identity labels to compute displacement vectors (Eqs. 3-5); the text should state explicitly that this analysis is post hoc and does not select dimensions in Eq. 2, to avoid implying a dimension-selection mechanism.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular reasoning found: the normalization is transductive postprocessing evaluated against baselines, not a fit to the retrieval labels or to the target metric.

full rationale

The paper's central operation, camera-specific feature normalization (Eq. 2), computes per-camera mean and standard deviation solely from unlabeled embedding features and known camera labels; identity labels are not used in estimating these statistics. The evaluation compares the same model with and without normalization on the same target set, which is transductive test-time adaptation rather than circular self-prediction: the transform is not fitted to mAP or Rank-1, and the tables report before/after numbers under identical model weights. The explanatory analysis in Sec. 4.2 is explicitly descriptive and post hoc; it uses variance of camera-mean features to inspect which dimensions dominate the centering effect and does not claim to predict held-out performance from statistics that encode the answer. The unsupervised learning contributions in Algorithm 1, debiased pseudo labeling and discarding single-camera clusters, are validated as ablations against the unmodified CC, PPLR, and PPLR-CAM baselines in Table 6 and Fig. 7, so the gains are not implied by construction. The paper also explicitly credits prior work for camera mean subtraction and camera-based batch normalization, so it is not renaming a known result as a new first-principles derivation. The Section J limitation mentions only computational cost, and the dependence on target camera labels and per-camera sample volume raised in Fig. 5 is a scope/assumption concern about deployment conditions, not a circularity in the derivation. No load-bearing self-citation chain appears: the references to prior camera-aware methods are external and are used as baselines or context, not as the justification for the paper's own empirical claims. Overall, the claimed analyses and improvements are self-contained experiments with direct baseline comparisons, so no circular step is exhibited.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The method introduces no new entities or fitted parameters in the sense of a learned model; it uses per-camera statistics estimated from unlabeled features. The key assumptions are availability of camera labels, sufficient samples per camera, and the validity of NMI as a bias measure. The DBSCAN epsilon and property-group count are experimental hyperparameters, not parameters fitted to the target result.

free parameters (2)
  • DBSCAN epsilon = 0.6
    Chosen for the main unsupervised learning experiments based on validation performance; a sweep in Appendix I shows the result depends on this value.
  • Number of property groups N = Varies (e.g., 50-100)
    Used in the property-specific normalization extension; the best group count is selected empirically from Figure 3.
assumptions (4)
  • domain assumption Camera labels of the test/target data are available at inference time.
    The camera-specific normalization in Eq. (2) requires the camera label y_i for each test sample, as stated in Section 4.1.
  • domain assumption NMI between cluster labels and camera labels is a valid measure of camera bias.
    Section 3 defines camera bias using NMI; the paper assumes this metric captures harmful feature-space separation rather than mere dataset label imbalance, although it does compare against ground-truth NMI values.
  • domain assumption The target set contains enough samples per camera (roughly 25 or more) to estimate reliable statistics.
    Figure 5 shows the method degrades with 5 samples per camera (7.6 mAP) and gains saturate around 100 samples, so the method depends on having sufficient per-camera data.
  • domain assumption A camera change moves features approximately as a translation in the embedding space.
    Section 4.2 analyzes displacement vectors and assumes consistent directional movement in camera-sensitive dimensions; this supports the explanation though the method itself is empirical.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Exploring the Camera Bias of Person Re-identification." pith.science (2026). https://pith.science/paper/QK355SET

@misc{pith2026250210195,
  author       = {Pith},
  title        = {Pith review of: Exploring the Camera Bias of Person Re-identification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QK355SET}},
  note         = {Machine review of arXiv:2502.10195}
}
read the original abstract

We empirically investigate the camera bias of person re-identification (ReID) models. Previously, camera-aware methods have been proposed to address this issue, but they are largely confined to training domains of the models. We measure the camera bias of ReID models on unseen domains and reveal that camera bias becomes more pronounced under data distribution shifts. As a debiasing method for unseen domain data, we revisit feature normalization on embedding vectors. While the normalization has been used as a straightforward solution, its underlying causes and broader applicability remain unexplored. We analyze why this simple method is effective at reducing bias and show that it can be applied to detailed bias factors such as low-level image properties and body angle. Furthermore, we validate its generalizability across various models and benchmarks, highlighting its potential as a simple yet effective test-time postprocessing method for ReID. In addition, we explore the inherent risk of camera bias in unsupervised learning of ReID models. The unsupervised models remain highly biased towards camera labels even for seen domain data, indicating substantial room for improvement. Based on observations of the negative impact of camera-biased pseudo labels on training, we suggest simple training strategies to mitigate the bias. By applying these strategies to existing unsupervised learning algorithms, we show that significant performance improvements can be achieved with minor modifications.

Figures

Figures reproduced from arXiv: 2502.10195 by the authors.

Figure 1
Figure 1. Cosine distance distributions of a camera-aware ReID model on (a) the training domain [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Analysis on the 384-dimensional embedding space of a ReID model. We measure the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Analysis on low-level properties. (a) Cosine similarity of displacement vectors by image [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (12 more)
Figure 4
Figure 4. Figure 4: Normalization result of Figure [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Results based on the number of sam￾ples used to calculate normalization parameters [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Risk of biased clusters. (a) Training results with varying pseudo label qualities for the same [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: Results of our training strategies for debiased unsupervised learning on MSMT17. [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Analysis on the 2048-dimensional embedding space of PPLR-CAM ( [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]
Figure 9
Figure 9. Figure 9: Analysis on the 1024-dimensional embedding space of SOLIDER ( [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]
Figure 10
Figure 10. Figure 10: Examples of applying the transformation functions to an image with four strength levels. [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: Statistics of low-level properties of images used in our experiments on each low-level [PITH_FULL_IMAGE:figures/full_fig_p016_11.png]
Figure 12
Figure 12. Figure 12: The Normalized Mutual Information (NMI) scores between property group labels and [PITH_FULL_IMAGE:figures/full_fig_p016_12.png]
Figure 13
Figure 13. Figure 13: Cosine similarity of displacement vectors of the features due to low-level transformations. [PITH_FULL_IMAGE:figures/full_fig_p016_13.png]
Figure 14
Figure 14. Figure 14: Templates of three body angle classes defined in our experiment. The blue and orange [PITH_FULL_IMAGE:figures/full_fig_p017_14.png]
Figure 15
Figure 15. Figure 15: The t-SNE result of features of PPLR-CAM trained on MSMT17 using samples from [PITH_FULL_IMAGE:figures/full_fig_p018_15.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Towards Anytime Retrieval: A Benchmark for Anytime Person Re-Identification

    cs.CV 2025-09 conditional novelty 6.0 of 10

    AT-USTC, a 403k-image RGB/IR dataset covering six time-based ReID scenarios, and Uni-AT, a multi-scenario model, are proposed, with Uni-AT achieving 55.8% any-time Rank-1 on the new benchmark.

  2. DART$^3$: Leveraging Distance for Test Time Adaptation in Person Re-Identification

    cs.CV 2025-05 conditional novelty 5.0 of 10

    A test-time adaptation method for person re-identification that learns per-camera scale and shift parameters with a top-k Euclidean distance objective, reducing camera bias without source data.

Reference graph

Works this paper leans on

52 extracted references · 42 canonical work pages · cited by 2 Pith papers

  1. [1]

    Ice: Inter-instance contrastive encoding for unsupervised person re-identification

    Hao Chen, Benoit Lagadec, and Francois Bremond. Ice: Inter-instance contrastive encoding for unsupervised person re-identification. In ICCV, 2021

    work page 2021

  2. [2]

    Beyond appearance: a semantic controllable self-supervised learning framework for human-centric visual tasks

    Weihua Chen, Xianzhe Xu, Jian Jia, Hao Luo, Yaohua Wang, Fan Wang, Rong Jin, and Xiuyu Sun. Beyond appearance: a semantic controllable self-supervised learning framework for human-centric visual tasks. In CVPR, 2023

    work page 2023

  3. [3]

    Ca-jaccard: Camera-aware jaccard distance for person re-identification

    Yiyu Chen, Zheyi Fan, Zhaoru Chen, and Yixuan Zhu. Ca-jaccard: Camera-aware jaccard distance for person re-identification. In CVPR, 2024

    work page 2024

  4. [4]

    Part-based pseudo label refinement for unsupervised person re-identification

    Yoonki Cho, Woo Jae Kim, Seunghoon Hong, and Sung-Eui Yoon. Part-based pseudo label refinement for unsupervised person re-identification. In CVPR, 2022

    work page 2022

  5. [5]

    Total recall: Automatic query expansion with a generative feature model for object retrieval

    Ondrej Chum, James Philbin, Josef Sivic, Michael Isard, and Andrew Zisserman. Total recall: Automatic query expansion with a generative feature model for object retrieval. In ICCV, 2007

    work page 2007

  6. [6]

    Generalizable person re-identification with relevance-aware mixture of experts

    Yongxing Dai, Xiaotong Li, Jun Liu, Zekun Tong, and Ling-Yu Duan. Generalizable person re-identification with relevance-aware mixture of experts. In CVPR, 2021

    work page 2021

  7. [7]

    Cluster contrast for unsupervised person re-identification

    Zuozhuo Dai, Guangyuan Wang, Weihao Yuan, Siyu Zhu, and Ping Tan. Cluster contrast for unsupervised person re-identification. In ACCV, 2022

    work page 2022

  8. [8]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at scale. ICLR, 2021

    2021

Show all 52 references
  1. [9]

    Identity-seeking self-supervised representation learning for generalizable person re-identification

    Zhaopeng Dou, Zhongdao Wang, Yali Li, and Shengjin Wang. Identity-seeking self-supervised representation learning for generalizable person re-identification. In ICCV, 2023

    2023

  2. [10]

    A density-based algorithm for discovering clusters in large spatial databases with noise

    Martin Ester, Hans-Peter Kriegel, J \"o rg Sander, Xiaowei Xu, et al. A density-based algorithm for discovering clusters in large spatial databases with noise. In KDD, 1996

    1996

  3. [11]

    Unsupervised person re-identification: Clustering and fine-tuning

    Hehe Fan, Liang Zheng, Chenggang Yan, and Yi Yang. Unsupervised person re-identification: Clustering and fine-tuning. ACM TOMM, 2018

    2018

  4. [12]

    Unsupervised pre-training for person re-identification

    Dengpan Fu, Dongdong Chen, Jianmin Bao, Hao Yang, Lu Yuan, Lei Zhang, Houqiang Li, and Dong Chen. Unsupervised pre-training for person re-identification. In CVPR, 2021

    2021

  5. [13]

    Large-scale pre-training for person re-identification with noisy labels

    Dengpan Fu, Dongdong Chen, Hao Yang, Jianmin Bao, Lu Yuan, Lei Zhang, Houqiang Li, Fang Wen, and Dong Chen. Large-scale pre-training for person re-identification with noisy labels. In CVPR, 2022

    2022

  6. [14]

    Self-paced contrastive learning with hybrid memory for domain adaptive object re-id

    Yixiao Ge, Feng Zhu, Dapeng Chen, Rui Zhao, et al. Self-paced contrastive learning with hybrid memory for domain adaptive object re-id. NeurIPS, 2020

    2020

  7. [15]

    End-to-end learning of deep visual representations for image retrieval

    Albert Gordo, Jon Almazan, Jerome Revaud, and Diane Larlus. End-to-end learning of deep visual representations for image retrieval. IJCV, 2017

    2017

  8. [16]

    1st place solution to visda-2020: Bias elimination for domain adaptive pedestrian re-identification

    Jianyang Gu, Hao Luo, Weihua Chen, Yiqi Jiang, Yuqi Zhang, Shuting He, Fan Wang, Hao Li, and Wei Jiang. 1st place solution to visda-2020: Bias elimination for domain adaptive pedestrian re-identification. arXiv preprint arXiv:2012.13498, 2020

    2020 arXiv

  9. [17]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In CVPR, 2016

    2016

  10. [18]

    Transreid: Transformer-based object re-identification

    Shuting He, Hao Luo, Pichao Wang, Fan Wang, Hao Li, and Wei Jiang. Transreid: Transformer-based object re-identification. In ICCV, 2021

    2021

  11. [19]

    In defense of the triplet loss for person re-identification

    Alexander Hermans, Lucas Beyer, and Bastian Leibe. In defense of the triplet loss for person re-identification. arXiv preprint arXiv:1703.07737, 2017

    2017 arXiv

  12. [20]

    Cloth-changing person re-identification from a single image with gait prediction and regularization

    Xin Jin, Tianyu He, Kecheng Zheng, Zhiheng Yin, Xu Shen, Zhen Huang, Ruoyu Feng, Jianqiang Huang, Zhibo Chen, and Xian-Sheng Hua. Cloth-changing person re-identification from a single image with gait prediction and regularization. In CVPR, 2022

    2022

  13. [21]

    Camera-driven representation learning for unsupervised domain adaptive person re-identification

    Geon Lee, Sanghoon Lee, Dohyung Kim, Younghoon Shin, Yongsang Yoon, and Bumsub Ham. Camera-driven representation learning for unsupervised domain adaptive person re-identification. In ICCV, 2023

    2023

  14. [22]

    Transmatcher: Deep image matching through transformers for generalizable person re-identification

    Shengcai Liao and Ling Shao. Transmatcher: Deep image matching through transformers for generalizable person re-identification. NeurIPS, 2021

    2021

  15. [23]

    A bottom-up clustering approach to unsupervised person re-identification

    Yutian Lin, Xuanyi Dong, Liang Zheng, Yan Yan, and Yi Yang. A bottom-up clustering approach to unsupervised person re-identification. In AAAI, 2019

    2019

  16. [24]

    Large-scale vehicle re-identification in urban surveillance videos

    Xinchen Liu, Wu Liu, Huadong Ma, and Huiyuan Fu. Large-scale vehicle re-identification in urban surveillance videos. In ICME, 2016

    2016

  17. [25]

    Mediapipe: A framework for building perception pipelines

    Camillo Lugaresi, Jiuqiang Tang, Hadon Nash, Chris McClanahan, Esha Uboweja, Michael Hays, Fan Zhang, Chuo-Ling Chang, Ming Guang Yong, Juhyun Lee, et al. Mediapipe: A framework for building perception pipelines. arXiv preprint arXiv:1906.08172, 2019

    1906 arXiv

  18. [26]

    Generalizing person re-identification by camera-aware invariance learning and cross-domain mixup

    Chuanchen Luo, Chunfeng Song, and Zhaoxiang Zhang. Generalizing person re-identification by camera-aware invariance learning and cross-domain mixup. In ECCV. Springer, 2020

    2020

  19. [27]

    Bag of tricks and a strong baseline for deep person re-identification

    Hao Luo, Youzhi Gu, Xingyu Liao, Shenqi Lai, and Wei Jiang. Bag of tricks and a strong baseline for deep person re-identification. In CVPR Workshops, 2019

    2019

  20. [28]

    An empirical study of vehicle re-identification on the ai city challenge

    Hao Luo, Weihua Chen, Xianzhe Xu, Jianyang Gu, Yuqi Zhang, Chong Liu, Yiqi Jiang, Shuting He, Fan Wang, and Hao Li. An empirical study of vehicle re-identification on the ai city challenge. In CVPR Workshops, 2021 a

    2021

  21. [29]

    Self-supervised pre-training for transformer-based person re-identification

    Hao Luo, Pichao Wang, Yi Xu, Feng Ding, Yanxin Zhou, Fan Wang, Hao Li, and Rong Jin. Self-supervised pre-training for transformer-based person re-identification. arXiv preprint arXiv:2111.12084, 2021 b

    2021 arXiv

  22. [30]

    Pose-guided feature alignment for occluded person re-identification

    Jiaxu Miao, Yu Wu, Ping Liu, Yuhang Ding, and Yi Yang. Pose-guided feature alignment for occluded person re-identification. In ICCV, 2019

    2019

  23. [31]

    Part-aware transformer for generalizable person re-identification

    Hao Ni, Yuke Li, Lianli Gao, Heng Tao Shen, and Jingkuan Song. Part-aware transformer for generalizable person re-identification. In ICCV, 2023

    2023

  24. [32]

    Maps of random walks on complex networks reveal community structure

    Martin Rosvall and Carl T Bergstrom. Maps of random walks on complex networks reveal community structure. PNAS, 2008

    2008

  25. [33]

    Body part-based representation learning for occluded person re-identification

    Vladimir Somers, Christophe De Vleeschouwer, and Alexandre Alahi. Body part-based representation learning for occluded person re-identification. In WACV, 2023

    2023

  26. [34]

    Generalizable person re-identification by domain-invariant mapping network

    Jifei Song, Yongxin Yang, Yi-Zhe Song, Tao Xiang, and Timothy M Hospedales. Generalizable person re-identification by domain-invariant mapping network. In CVPR, 2019

    2019

  27. [35]

    Dissecting person re-identification from the viewpoint of viewpoint

    Xiaoxiao Sun and Liang Zheng. Dissecting person re-identification from the viewpoint of viewpoint. In CVPR, 2019

    2019

  28. [36]

    Camera-aware proxies for unsupervised person re-identification

    Menglin Wang, Baisheng Lai, Jianqiang Huang, Xiaojin Gong, and Xian-Sheng Hua. Camera-aware proxies for unsupervised person re-identification. In AAAI, 2021

    2021

  29. [37]

    Person transfer gan to bridge domain gap for person re-identification

    Longhui Wei, Shiliang Zhang, Wen Gao, and Qi Tian. Person transfer gan to bridge domain gap for person re-identification. In CVPR, 2018

    2018

  30. [38]

    Rgb-infrared cross-modality person re-identification

    Ancong Wu, Wei-Shi Zheng, Hong-Xing Yu, Shaogang Gong, and Jianhuang Lai. Rgb-infrared cross-modality person re-identification. In ICCV, 2017

    2017

  31. [39]

    Unsupervised visible-infrared person re-identification via progressive graph matching and alternate learning

    Zesen Wu and Mang Ye. Unsupervised visible-infrared person re-identification via progressive graph matching and alternate learning. In CVPR, 2023

    2023

  32. [40]

    Deep learning for person re-identification: A survey and outlook

    Mang Ye, Jianbing Shen, Gaojie Lin, Tao Xiang, Ling Shao, and Steven CH Hoi. Deep learning for person re-identification: A survey and outlook. TPAMI, 2021

    2021

  33. [41]

    Unsupervised person re-identification by soft multilabel learning

    Hong-Xing Yu, Wei-Shi Zheng, Ancong Wu, Xiaowei Guo, Shaogang Gong, and Jian-Huang Lai. Unsupervised person re-identification by soft multilabel learning. In CVPR, 2019

    2019

  34. [42]

    Unsupervised domain adaptation for person re-identification via heterogeneous graph alignment

    Minying Zhang, Kai Liu, Yidong Li, Shihui Guo, Hongtao Duan, Yimin Long, and Yi Jin. Unsupervised domain adaptation for person re-identification via heterogeneous graph alignment. In AAAI, 2021

    2021

  35. [43]

    Self-training with progressive augmentation for unsupervised cross-domain person re-identification

    Xinyu Zhang, Jiewei Cao, Chunhua Shen, and Mingyu You. Self-training with progressive augmentation for unsupervised cross-domain person re-identification. In ICCV, 2019

    2019

  36. [44]

    Scalable person re-identification: A benchmark

    Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jingdong Wang, and Qi Tian. Scalable person re-identification: A benchmark. In ICCV, 2015

    2015

  37. [45]

    A discriminatively learned cnn embedding for person reidentification

    Zhedong Zheng, Liang Zheng, and Yi Yang. A discriminatively learned cnn embedding for person reidentification. ACM TOMM, 2017

    2017

  38. [46]

    Re-ranking person re-identification with k-reciprocal encoding

    Zhun Zhong, Liang Zheng, Donglin Cao, and Shaozi Li. Re-ranking person re-identification with k-reciprocal encoding. In CVPR, 2017 a

    2017

  39. [47]

    Re-ranking person re-identification with k-reciprocal encoding

    Zhun Zhong, Liang Zheng, Donglin Cao, and Shaozi Li. Re-ranking person re-identification with k-reciprocal encoding. In CVPR, 2017 b

    2017

  40. [48]

    Rethinking the distribution gap of person re-identification with camera-based batch normalization

    Zijie Zhuang, Longhui Wei, Lingxi Xie, Tianyu Zhang, Hengheng Zhang, Haozhe Wu, Haizhou Ai, and Qi Tian. Rethinking the distribution gap of person re-identification with camera-based batch normalization. In ECCV, 2020

    2020

  41. [49]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

  42. [50]

    @esa (Ref

    \@ifxundefined[1] #1\@undefined \@firstoftwo \@secondoftwo \@ifnum[1] #1 \@firstoftwo \@secondoftwo \@ifx[1] #1 \@firstoftwo \@secondoftwo [2] @ #1 \@temptokena #2 #1 @ \@temptokena \@ifclassloaded agu2001 natbib The agu2001 class already includes natbib coding, so you should ...

  43. [51]

    \@lbibitem[] @bibitem@first@sw\@secondoftwo \@lbibitem[#1]#2 \@extra@b@citeb \@ifundefined br@#2\@extra@b@citeb \@namedef br@#2 \@nameuse br@#2\@extra@b@citeb \@ifundefined b@#2\@extra@b@citeb @num @parse #2 @tmp #1 NAT@b@open@#2 NAT@b@shut@#2 \@ifnum @merge>\@ne @bibitem@firs...

  44. [52]

    We argue that the existing USL algorithms have two limitations introducing the camera bias into the models

    @open @close @open @close and [1] URL: #1 \@ifundefined chapter * \@mkboth \@ifxundefined @sectionbib * \@mkboth * \@mkboth\@gobbletwo \@ifclassloaded amsart * \@ifclassloaded amsbook * \@ifxundefined @heading @heading NAT@ctr thebibliography [1] @ \@biblabel @NAT@ctr \@bibset...

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.