Pith. sign in

REVIEW 3 major objections 5 minor 38 references

Background Matters Too: A Language-Enhanced Adversarial Framework for Person Re-Identification

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Jointly modeling foreground and background language cues with a diversity loss improves person re-identification and matches or beats prior state of the art on four benchmarks.

desk verdict FBA is a plausible but under-verified extension of CLIP-based ReID; the background-caption idea is worth engaging, but the caption-quality assumption and a contradictory pooling description keep the evidence conditional. read the letter →

arxiv 2509.03032 v1 pith:XPIZWTWH submitted 2025-09-03 cs.CV

Kaicong Huang , Talha Azfar , Jack M. Reilly , Thomas Guggisberg , Ruimin Ke This is my paper · ORCID
classification cs.CV
keywords personre-identificationvision-languagemodelsforeground-backgroundsemanticscross-modalattentiondiversitylossocclusionrobustnessCLIPmapdifferentialpooling
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

This paper tries to establish that background semantics are not noise to be discarded but a usable signal for person re-identification. The proposed FBA framework models foreground and background as two semantic domains, aligns image patches and text tokens within each domain, and adversarially pushes the domains apart so the network suppresses distractors while attending to identity cues. On the paper's reported numbers, this matches or surpasses prior state-of-the-art results on two holistic benchmarks (DukeMTMC-reID, CUHK03-NP) and two occluded benchmarks (Occluded-Duke, Occluded-ReID), with the largest gains on occluded data. If the result holds, background-aware language guidance would give ReID models a cheap, annotation-free route to occlusion robustness.

What carries the argument

The machinery has three linked pieces. First, a dual-branch cross-modal attention module runs four weight-shared transformer blocks (two per branch), using image patches as query, key, and value together with text tokens so both modalities attend to each other; one branch handles foreground captions and the other background captions. Second, a diversity loss combines a multi-view triplet term over foreground/background visual-text features with a contrastive term that pulls paired visual and text features together within the same semantic region, so the two domains separate while intra-domain modalities align. Third, an attention map differential pooling strategy computes cosine similarity between the foreground and background cross-attention maps, normalizes it into a token mask, and uses it to aggregate identity-relevant text-guided patch embeddings instead of only the CLS and EOS tokens.

What would settle it

Generate captions for a fixed subset of CUHK03-NP or Occluded-Duke, have annotators label each caption as a correct or incorrect foreground or background description, and retrain FBA with only correct captions versus shuffled or corrupted captions; if the mAP and Rank-1 gap between FBA and its foreground-only baseline persists under corrupted captions, the diversity-loss improvement is not caused by accurate background semantics.

Watch

Extended reading notes

Core claim

The paper's central claim is that jointly modeling foreground and background semantic information improves person re-identification beyond what foreground-only CLIP-style alignment achieves. FBA uses a large language model to produce separate foreground and background captions for each image, feeds visual patches and text tokens through dual weight-shared cross-modal branches, and optimizes with identity and triplet losses plus a diversity loss that aligns same-domain features while driving foreground and background features apart. The reported results are the evidence: on CUHK03-NP FBA reaches 85.3 mAP and 86.6 Rank-1 versus 83.1 and 85.1 for the prior PromptSG method; on Occluded-Duke it reaches 60.5 mAP and 69.5 Rank-1, and it transfers to the test-only Occluded-ReID set with 84.0 mAP and 85.4 Rank-1. The paper interprets these numbers as showing that background semantics help the model separate target from distractor in both holistic and occluded settings, without explicit occlusion modules or part annotations.

Load-bearing premise

The load-bearing premise is that the LLM-generated foreground and background captions are accurate enough to supervise region separation; the paper provides no manual evaluation, caption statistics, or qualitative checks, so if the captions describe the wrong region or miss key objects, the reported gains could come from caption artifacts rather than genuine foreground/background discrimination.

Editorial extensions

If this is right

  • If background captions are an effective training signal, ReID models can exploit semantic context without manual part or foreground masks.
  • Occlusion robustness can be obtained from foreground/background separation losses rather than dedicated occlusion modules, simplifying the pipeline.
  • Because the text encoder and the caption generator are frozen, the added capacity is small and the framework can be applied to existing CLIP-based ReID backbones.
  • The attention differential pooling strategy could generalize to any two mutually exclusive semantic groups, not just persons.

Reading between the lines

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

  • The method's gains likely depend on caption quality, so a natural next experiment would freeze the architecture and swap caption sources; large drops would localize the contribution to language supervision.
  • Background captions may be most valuable in open-world settings where test-time distractors are unseen, because the background text supplies a semantic negative that generalizes beyond dataset-specific noise.
  • Prompt wording is under-explored in the paper; the same dual-branch design with prompts targeted at occluder categories such as vehicles, railings, and other pedestrians might transfer better to surveillance scenes.
  • Caption-image alignment could double as a diagnosis tool: the attention map difference already highlights which text tokens separate foreground from background, which could be used to detect when generated captions mislabel a region.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper proposes FBA, a CLIP-based person re-identification framework that explicitly models both foreground and background semantics. LLaVA v1.5-7b generates foreground and background captions; a frozen CLIP text encoder and a trainable ViT visual encoder feed a dual-branch cross-modal attention module with four weight-shared blocks. An attention-map differential pooling strategy aggregates token embeddings, and the model is trained with identity loss, triplet loss, and a diversity loss that combines an intra-semantic contrastive term and an inter-semantic triplet term. Experiments on DukeMTMC-reID, CUHK03-NP, Occluded-Duke, and Occluded-ReID report results that match or exceed several prior methods, with ablations on two datasets. The central claim is that jointly modeling foreground and background semantics improves ReID accuracy, especially under occlusion.

Significance. If the central claim holds, FBA is a worthwhile contribution to language-guided ReID: it treats background captions as a complementary signal rather than discarding them, and the experimental section consistently improves over the authors' own baseline across four datasets. The ablation isolating the cross-modal interaction, the diversity loss, and the pooling mask is useful, and the use of LLaVA-generated captions avoids manual attribute annotation. However, the validity of the reported gains rests on caption quality and on the correct interpretation of the pooling mechanism, and the single-run point estimates without released artifacts leave the quantitative claims insufficiently supported. These are addressable in revision.

major comments (3)
  1. [Multimodal Representation Encoding (LLaVA captions) and Objective Function (Eqs. 6-7)] The central claim depends on LLaVA v1.5-7b generated foreground and background captions being region-correct, but the paper reports no validation of these captions: no manual evaluation, caption statistics, or quality checks, and the captions are not released. The foreground prompt asks LLaVA to "Describe the appearance of persons in the image" in plural, so in multi-person or occluded images (Occluded-Duke, Occluded-ReID) the model may describe a bystander or the occluder rather than the target identity, and the background prompt may include the target. Since the intra-semantic contrastive loss Lcon (Eq. 7) and the inter-semantic diversity loss Ltri-div (Eq. 6) are computed directly against these captions, wrong-region captions would make the reported gains an artifact of caption noise or added cross-modal parameters rather than evidence that background semantics help. Please provide caption-quality evaluation on a sample of each dataset, release the captions, and include an ablation that replaces LLaVA captions with random or mismatched captions to show the signal is real.
  2. [Attention Map Differential Pooling (Eqs. 2-3)] The mechanism as written appears to do the opposite of what the text claims. The text says "To apply more attention to tokens that better separate foreground and background," but Eq. (2) computes the cosine similarity between the foreground and background attention columns, and Eq. (3) min-max normalizes it so that high similarity receives the largest mask weight. A strategy that rewards separation should up-weight dissimilar columns, not similar ones. Because the ablation in Table 3 attributes a gain to this mask, the inconsistency matters for understanding the method. Please clarify whether high similarity is the intended pooling criterion and, if so, explain why similar foreground/background attention patterns correspond to separation; otherwise revise the formulation.
  3. [Experiments: Tables 1 and 3] All reported numbers are single-run point estimates without error bars or significance tests. Several key margins are small: on DukeMTMC the mAP advantage over PromptSG is +0.1, and in Table 3 the diversity-loss row gains 0.7 mAP on DukeMTMC and 1.8 mAP on CUHK03-NP. With single runs it is not possible to tell whether these differences are real or due to training stochasticity. Additionally, the "Baseline" row in Table 1 is not defined in the implementation details (which visual encoder, which loss combination, whether it includes the frozen text encoder). Please report mean and standard deviation over at least three runs for the main tables and define the baseline configuration precisely.
minor comments (5)
  1. [Table 2] The dataset name is misspelled as "CHUK03-NP"; it should be "CUHK03-NP".
  2. [Objective Function, Eq. (9)] The set P and the phrase "all possible pairs will be calculated twice" are not formally defined; please specify how the triplet arguments (a, b, c, d) are constructed from the four feature types and clarify the "twice" statement.
  3. [Multimodal Representation Encoding] The foreground prompt is written in the plural ("persons"), which is inconsistent with the single-target ReID setting; consider using the singular "person" and describing how images with multiple people are handled.
  4. [Figure 2] The t-SNE visualization is qualitative; a quantitative measure of foreground/background feature separation (for example, average inter-similarity versus intra-similarity) would support the claim.
  5. [Throughout] The term "adversarial" is used for a diversity loss that is not adversarial in the minimax sense; consider renaming it or adding an explanation of the intended meaning.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: FBA's benchmark results are external and its losses, captions, and hyperparameters are not constructed from the reported metrics.

full rationale

The paper's claimed derivation chain is the FBA framework: LLaVA-generated foreground/background captions are encoded by a frozen CLIP text encoder, a dual-branch cross-modal attention module mixes patch and token embeddings, and the losses (L_ID, L_Triplet, L_con, L_tri-div) enforce identity discrimination and foreground/background separation. None of these quantities is defined from the benchmark outputs it later predicts; the reported mAP/R-1 values on DukeMTMC, CUHK03-NP, Occluded-Duke, and Occluded-ReID come from standard external evaluation protocols, and the comparison baselines are independently published methods. The one self-citation (TransitReID) is unrelated to the proposed losses and is not load-bearing. The per-dataset learning rates in the implementation details are ordinary hyperparameter tuning on the same benchmarks, not a fitted parameter renamed as a prediction, and the paper's own ablated baseline (L_ID+L_Triplet) makes the incremental gains empirical rather than tautological. The LLaVA caption-quality concern and the wording of the attention-map differential pooling strategy are potential correctness or interpretability issues, but they do not constitute circularity: the captions are external inputs, and the pooling mask is not set equal to any loss or metric. The paper itself notes the prompts 'are just simply selected and still need further exploration,' an honest limitation rather than a circular step. Overall, the central claim is supported by external benchmarks and by ablations, so no circularity is present.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The central claim rests on external pretrained models (CLIP and LLaVA) whose outputs are not independently verified, and on several hand-picked hyperparameters tuned on the same benchmarks used for evaluation. No new physical or conceptual entity is introduced.

free parameters (4)
  • Diversity loss weight lambda = 0.5
    Balance factor in Eq. (10); no sensitivity analysis is shown, and it directly controls the strength of the proposed loss.
  • Triplet margin m = 0.3
    Standard margin in Eq. (5); chosen by hand without reported tuning or sensitivity analysis.
  • Per-dataset base learning rate = 8e-5 for DukeMTMC and Occluded-Duke, 1.2e-4 for CUHK03-NP
    The authors state the base learning rate is empirically set per dataset to improve training stability; this is tuning on the evaluation benchmarks.
  • Training resolution and stride = 384x128 pixels, stride 12
    Selected after the ablation in Table 4; affects the final reported numbers and is a hand-chosen configuration.
assumptions (3)
  • domain assumption CLIP ViT-B/16 visual encoder and frozen CLIP text encoder provide useful, composable semantic representations for person images.
    The framework inherits CLIP's pretrained alignment and assumes it transfers to the ReID domain; no direct validation of this transfer is provided beyond the final benchmark numbers.
  • domain assumption LLaVA v1.5-7b generated foreground and background captions are semantically accurate enough to supervise region separation.
    Captions are generated without manual validation, caption statistics, or quality checks, even though they are the sole source of semantic supervision for the background branch.
  • domain assumption mAP and Rank-1 on DukeMTMC, CUHK03-NP, Occluded-Duke, and Occluded-ReID are accepted measures of person re-identification performance.
    Standard metrics in the field, but the paper reports no error bars and does not establish that differences of about one point are significant.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Background Matters Too: A Language-Enhanced Adversarial Framework for Person Re-Identification." pith.science (2026). https://pith.science/paper/XPIZWTWH

@misc{pith2026250903032,
  author       = {Pith},
  title        = {Pith review of: Background Matters Too: A Language-Enhanced Adversarial Framework for Person Re-Identification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XPIZWTWH}},
  note         = {Machine review of arXiv:2509.03032}
}
read the original abstract

Person re-identification faces two core challenges: precisely locating the foreground target while suppressing background noise and extracting fine-grained features from the target region. Numerous visual-only approaches address these issues by partitioning an image and applying attention modules, yet they rely on costly manual annotations and struggle with complex occlusions. Recent multimodal methods, motivated by CLIP, introduce semantic cues to guide visual understanding. However, they focus solely on foreground information, but overlook the potential value of background cues. Inspired by human perception, we argue that background semantics are as important as the foreground semantics in ReID, as humans tend to eliminate background distractions while focusing on target appearance. Therefore, this paper proposes an end-to-end framework that jointly models foreground and background information within a dual-branch cross-modal feature extraction pipeline. To help the network distinguish between the two domains, we propose an intra-semantic alignment and inter-semantic adversarial learning strategy. Specifically, we align visual and textual features that share the same semantics across domains, while simultaneously penalizing similarity between foreground and background features to enhance the network's discriminative power. This strategy drives the model to actively suppress noisy background regions and enhance attention toward identity-relevant foreground cues. Comprehensive experiments on two holistic and two occluded ReID benchmarks demonstrate the effectiveness and generality of the proposed method, with results that match or surpass those of current state-of-the-art approaches.

Figures

Figures reproduced from arXiv: 2509.03032 by the authors.

Figure 1
Figure 1. Frameworks of different person re-identification [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. t-SNE visualization of cross-modal features, where [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Overview of the proposed Foreground and Background Adversarial Image-text Person Re-identification framework [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Attention map visualization of FBA on Occluded [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 24 canonical work pages

  1. [1]

    Gao, L.; Zhang, H.; Gao, Z.; Guan, W.; Cheng, Z.; and Wang, M. 2020. Texture semantically aligned with visibility-aware for partial person re-identification. In Proceedings of the 28th ACM international conference on multimedia, 3771--3779

    work page 2020

  2. [2]

    He, S.; Luo, H.; Wang, P.; Wang, F.; Li, H.; and Jiang, W. 2021. Transreid: Transformer-based object re-identification. In Proceedings of the IEEE/CVF international conference on computer vision, 15013--15022

    2021

  3. [3]

    Huang, K.; Azfar, T.; Reilly, J.; and Ke, R. 2025. TransitReID: Transit OD Data Collection with Occlusion-Resistant Dynamic Passenger Re-Identification. arXiv preprint arXiv:2504.11500

    arXiv 2025

  4. [4]

    Jiang, D.; and Ye, M. 2023. Cross-modal implicit relation reasoning and aligning for text-to-image person retrieval. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2787--2797

    2023

  5. [5]

    Kang, R.; Song, Y.; Gkioxari, G.; and Perona, P. 2025. Is CLIP ideal? No. Can we fix it? Yes! arXiv preprint arXiv:2503.08723

    arXiv 2025

  6. [6]

    Kim, M.; Cho, M.; Lee, H.; Cho, S.; and Lee, S. 2022. Occluded person re-identification via relational adaptive feature correction learning. In ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2719--2723. IEEE

    work page 2022

  7. [7]

    Li, S.; Sun, L.; and Li, Q. 2023. Clip-reid: exploiting vision-language model for image re-identification without concrete text labels. In Proceedings of the AAAI conference on artificial intelligence, volume 37, 1405--1413

    work page 2023

  8. [8]

    Li, W.; Zhao, R.; Xiao, T.; and Wang, X. 2014. Deepreid: Deep filter pairing neural network for person re-identification. In Proceedings of the IEEE conference on computer vision and pattern recognition, 152--159

    2014

Show all 38 references
  1. [9]

    Li, Y.; He, J.; Zhang, T.; Liu, X.; Zhang, Y.; and Wu, F. 2021. Diverse part discovery: Occluded person re-identification with part-aware transformer. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2898--2907

    2021

  2. [10]

    Liu, D.; Han, S.; Chen, Y.; Xia, C.; and Zhao, J. 2022. Foreground-guided textural-focused person re-identification. Neurocomputing, 483: 235--248

    2022

  3. [11]

    Liu, H.; Li, C.; Wu, Q.; and Lee, Y. J. 2023. Visual Instruction Tuning

    2023

  4. [12]

    Liu, Y.; Zhou, W.; Liu, J.; Qi, G.-J.; Tian, Q.; and Li, H. 2021. An end-to-end foreground-aware network for person re-identification. IEEE Transactions on Image Processing, 30: 2060--2071

    2021

  5. [13]

    Mao, J.; Yao, Y.; Sun, Z.; Huang, X.; Shen, F.; and Shen, H.-T. 2023. Attention map guided transformer pruning for occluded person re-identification on edge device. IEEE Transactions on Multimedia, 25: 1592--1599

    2023

  6. [14]

    Miao, J.; Wu, Y.; Liu, P.; Ding, Y.; and Yang, Y. 2019. Pose-guided feature alignment for occluded person re-identification. In Proceedings of the IEEE/CVF international conference on computer vision, 542--551

    2019

  7. [15]

    Ning, E.; Wang, C.; Zhang, H.; Ning, X.; and Tiwari, P. 2024. Occluded person re-identification with deep learning: a survey and perspectives. Expert systems with applications, 239: 122419

    2024

  8. [16]

    W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al

    Radford, A.; Kim, J. W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al. 2021. Learning transferable visual models from natural language supervision. In International conference on machine learning, 8748--8763. PmLR

    2021

  9. [17]

    Sun, Y.; Zheng, L.; Yang, Y.; Tian, Q.; and Wang, S. 2018. Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline). In Proceedings of the European conference on computer vision (ECCV), 480--496

    2018

  10. [18]

    Tan, L.; Dai, P.; Ji, R.; and Wu, Y. 2022. Dynamic prototype mask for occluded person re-identification. In Proceedings of the 30th ACM international conference on multimedia, 531--540

    2022

  11. [19]

    Tan, L.; Xia, J.; Liu, W.; Dai, P.; Wu, Y.; and Cao, L. 2024. Occluded person re-identification via saliency-guided patch transfer. In Proceedings of the AAAI conference on artificial intelligence, volume 38, 5070--5078

    2024

  12. [20]

    Wang, G.; Yang, S.; Liu, H.; Wang, Z.; Yang, Y.; Wang, S.; Yu, G.; Zhou, E.; and Sun, J. 2020. High-order information matters: Learning relation and topology for occluded person re-identification. In Proceedings of the IEEE/CVF conference on computer vision and pattern recogni...

    2020

  13. [21]

    Wang, G.; Yu, F.; Li, J.; Jia, Q.; and Ding, S. 2023. Exploiting the textual potential from vision-language pre-training for text-based person search. arXiv preprint arXiv:2303.04497

    2023 arXiv

  14. [22]

    Wang, P.; Zhao, Z.; Su, F.; and Meng, H. 2022 a . Ltreid: Factorizable feature generation with independent components for long-tailed person re-identification. IEEE Transactions on Multimedia, 25: 4610--4622

    2022

  15. [23]

    Wang, Z.; Zhu, F.; Tang, S.; Zhao, R.; He, L.; and Song, J. 2022 b . Feature erasing and diffusion network for occluded person re-identification. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 4754--4763

    2022

  16. [24]

    Wu, X.; Ma, W.; Guo, D.; Zhou, T.; Zhao, S.; and Cai, Z. 2024. Text-based occluded person re-identification via multi-granularity contrastive consistency learning. In Proceedings of the AAAI conference on artificial intelligence, volume 38, 6162--6170

    2024

  17. [25]

    Xia, J.; Tan, L.; Dai, P.; Zhao, M.; Wu, Y.; and Cao, L. 2024. Attention disturbance and dual-path constraint network for occluded person re-identification. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, 6198--6206

    2024

  18. [26]

    Yan, S.; Dong, N.; Zhang, L.; and Tang, J. 2023. Clip-driven fine-grained text-image person re-identification. IEEE Transactions on Image Processing, 32: 6032--6046

    2023

  19. [27]

    Yang, D.; He, J.; Ma, Y.; Yu, Q.; and Zhang, T. 2023. Foreground-background distribution modeling transformer for visual object tracking. In Proceedings of the IEEE/CVF international conference on computer vision, 10117--10127

    2023

  20. [28]

    Yang, S.; and Zhang, Y. 2024. MLLMReID: multimodal large language model-based person re-identification. arXiv preprint arXiv:2401.13201

    2024 arXiv

  21. [29]

    Yang, Z.; Wu, D.; Wu, C.; Lin, Z.; Gu, J.; and Wang, W. 2024. A pedestrian is worth one prompt: Towards language guidance person re-identification. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 17343--17353

    2024

  22. [30]

    Zheng, Z.; Zheng, L.; and Yang, Y. 2017. Unlabeled samples generated by gan improve the person re-identification baseline in vitro. In Proceedings of the IEEE international conference on computer vision, 3754--3762

    2017

  23. [31]

    Zhong, Z.; Zheng, L.; Kang, G.; Li, S.; and Yang, Y. 2020. Random erasing data augmentation. In Proceedings of the AAAI conference on artificial intelligence, volume 34, 13001--13008

    2020

  24. [32]

    C.; and Liu, Z

    Zhou, K.; Yang, J.; Loy, C. C.; and Liu, Z. 2022. Learning to prompt for vision-language models. International Journal of Computer Vision, 130(9): 2337--2348

    2022

  25. [33]

    Zhu, H.; Ke, W.; Li, D.; Liu, J.; Tian, L.; and Shan, Y. 2022. Dual cross-attention learning for fine-grained visual categorization and object re-identification. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 4692--4702

    2022

  26. [34]

    Zhu, K.; Guo, H.; Liu, Z.; Tang, M.; and Wang, J. 2020. Identity-guided human semantic parsing for person re-identification. In Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part III 16, 346--363. Springer

    2020

  27. [35]

    Zhu, K.; Guo, H.; Zhang, S.; Wang, Y.; Liu, J.; Wang, J.; and Tang, M. 2023. Aaformer: Auto-aligned transformer for person re-identification. IEEE Transactions on Neural Networks and Learning Systems

    2023

  28. [36]

    Zhuo, J.; Chen, Z.; Lai, J.; and Wang, G. 2018. Occluded person re-identification. In 2018 IEEE international conference on multimedia and expo (ICME), 1--6. IEEE

    2018

  29. [37]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  30. [38]

    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...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.