Pith. sign in

REVIEW 2 major objections 6 minor 74 references

ABD-Net: Attentive but Diverse Person Re-Identification

T0 review · 2 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read ABD-Net claims that attention and diversity are complementary, and that combining channel/position attention with orthogonality regularization sets state-of-the-art results on three person re-identification benchmarks.

desk verdict Solid empirical Re-ID paper with a borrowed attention recipe and a new orthogonality regularizer whose stated math does not match the implementation; results likely hold, but the paper overstates the mechanism. read the letter →

arxiv 1908.01114 v3 pith:5DAIPN7E submitted 2019-08-03 cs.CV

Tianlong Chen , Shaojin Ding , Jingyi Xie , Ye Yuan , Wuyang Chen , Yang Yang , Zhou Ren , Zhangyang Wang This is my paper · ORCID
classification cs.CV
keywords personre-identificationattentionmechanismorthogonalityregularizationfeaturediversitychannelmodulepositionspectralvaluedifferencedeepmetriclearning
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 argues that attention mechanisms alone are not enough for person re-identification: they focus features on informative regions but also make those features correlated and redundant, which hurts Euclidean-distance matching. To fix this, ABD-Net combines two attention modules (one for channels, one for spatial positions) with an orthogonality regularizer that encourages the hidden activations and weights to be diverse and uncorrelated. On three popular benchmarks the paper reports new state-of-the-art results, with mAP of 88.28% on Market-1501, 78.59% on DukeMTMC-Re-ID, and 60.80% on MSMT17. The paper's message is that diversity should be treated as a necessary complement to attention in retrieval models.

What carries the argument

The central mechanism is the spectral value difference orthogonality (SVDO) regularizer, which approximates the condition number of a feature or weight matrix by penalizing $\|\lambda_1(FF^T) - \lambda_2(FF^T)\|_2^2$, the difference between the largest and smallest eigenvalues of the Gram matrix $FF^T$, estimated with only two power iterations. Complementing it are the Channel Attention Module (CAM) and the Position Attention Module (PAM), which compute affinity matrices over channels and over pixel positions, respectively, to re-weight the feature maps. The two-branch architecture keeps a global branch alongside an attentive branch, so the network preserves global context while learning focused, decorrelated attentive features; the paper's argument rests on the interaction of these parts.

What would settle it

Compute the actual condition numbers of the Gram matrices encountered while training ABD-Net and compare them with the SVDO penalty: if matrices with low SVDO penalty still have high condition numbers, then SVDO is not enforcing the stated orthogonality. A second test is to replace SVDO with an exact or better-approximated condition-number or spectral-norm regularizer and check whether the reported gains on Market-1501, DukeMTMC-Re-ID, and MSMT17 persist.

Watch

Extended reading notes

Core claim

ABD-Net's central claim is that attentive and diverse feature embeddings are complementary and jointly needed for strong person re-identification. The network adds a Channel Attention Module (CAM) and a Position Attention Module (PAM) to aggregate semantically related channels and spatial locations, and it applies a spectral value difference orthogonality (SVDO) regularizer to both hidden activations and weights. The SVDO term penalizes the difference between the largest and smallest eigenvalues of the Gram matrix $FF^T$, aiming to keep the feature channels decorrelated even while attention concentrates them on the person. In the paper's experiments, each of these components contributes a measurable gain, and the full model outperforms prior state-of-the-art methods on Market-1501, DukeMTMC-Re-ID, and MSMT17, with the largest margins on the two larger datasets.

Load-bearing premise

The load-bearing assumption is that penalizing the difference between the largest and smallest eigenvalues of the Gram matrix, as SVDO does, faithfully enforces the condition-number orthogonality described in the paper; the difference form is chosen to avoid dividing by a small eigenvalue, but a matrix with eigenvalues 0.01 and 1.01 has a small penalty and a condition number of 101, so the two are not equivalent.

Editorial extensions

If this is right

  • On all three benchmarks, ABD-Net's mAP exceeds every prior method the paper lists: 88.28% on Market-1501, 78.59% on DukeMTMC-Re-ID, and 60.80% on MSMT17.
  • The ablation study shows each component (CAM, PAM, feature orthogonality, weight orthogonality) adds a positive increment, and the increments combine rather than cancel.
  • The correlation analysis shows attention alone raises average off-diagonal channel correlation (0.368 in the displayed example) and the orthogonality regularizer reduces it (0.214), consistent with the paper's causal story.
  • Because SVDO uses only two power iterations and is applied to activations and weights during training, it adds no extra computation at inference time, making the recipe practical for large-scale Re-ID.

Reading between the lines

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

  • The paper's stated rationale for SVDO (penalizing condition number) is weaker than the implementation, since penalizing the difference between extreme eigenvalues does not control their ratio when the smallest eigenvalue is tiny; if that gap occurs in practice, the diversity gains may partly come from a different mechanism, such as increased regularization in general.
  • The same attention-plus-diversity recipe could plausibly transfer to other Euclidean-distance retrieval tasks, such as face or vehicle re-identification, which share the tendency of attention to collapse features onto a compact subspace.
  • A controlled comparison that replaces SVDO with an exact or better-approximated condition-number penalty would separate the mathematical claim from the empirical recipe and would test whether two power iterations are sufficient.
  • The two-branch architecture (global plus attentive, with concatenated outputs) resembles other multi-branch Re-ID designs; an ablation that adds the second branch without CAM/PAM would isolate the specific contribution of the attention and diversity terms.
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

2 major / 6 minor

Summary. The paper proposes ABD-Net, a person re-identification model that integrates two attention modules (channel attention and position attention) with a new orthogonality regularizer, SVDO, applied to both feature activations and network weights. The authors claim that this combination yields features that are attentive, diverse, and discriminative, and report state-of-the-art results on Market-1501, DukeMTMC-Re-ID, and MSMT17. The paper includes extensive ablations, visualizations of attention maps and feature correlation, and provides public code.

Significance. If the claims are validated, the paper makes a useful empirical contribution by demonstrating that attention and diversity regularization can be combined effectively in a Re-ID system, and it provides a strong baseline on three benchmarks. The strengths are the reproducible experimental pipeline, detailed ablations, and the public code release. However, the paper's central theoretical claim about the SVDO regularizer—that it directly constrains the condition number of the Gram matrix—is not supported as written. The mismatch between the condition-number objective in Eq. (3) and the spectral-difference objective in Eq. (4) is substantive because the regularizer is the paper's main novelty and Table 1 attributes a 4.7 mAP gain to the O.F. term. The significance is therefore conditional on the authors correcting or re-framing the theoretical derivation.

major comments (2)
  1. [Section 3.2, Eq. (5)] The paper claims that Eq. (4), β||λ1(FFᵀ) − λ2(FFᵀ)||², is a conversion of the condition-number penalty in Eq. (3), β||κ(F) − 1||², but the two objectives are not equivalent. Equation (3) is scale-invariant, penalizing (σ_max/σ_min − 1)², while Eq. (4) is scale-sensitive, penalizing (σ_max² − σ_min²)². For example, a Gram matrix with eigenvalues 0.01 and 1.01 has condition number about 10 but penalty 1, while scaling all eigenvalues by 100 leaves the condition number unchanged but multiplies the penalty by 10⁴. Thus the model can reduce the Eq. (4) penalty by shrinking feature norms rather than by decorrelating features. The footnote after Eq. (4) acknowledges that the difference is intentional for numerical stability, but the main text says "convert" and the contribution bullet says SVDO "directly constrains the conditional number"; this is contradictory and misleading. This is load-bearing because the diversity mechanism's theoretical justification rests on condition-number control, and the paper does not provide evidence that the difference penalty behaves like the condition number in practice. Please revise Section 3.2 to state precisely what SVDO penalizes, or modify the loss to be scale-invariant (e.g., penalizing (λ1 − λmin)/λ1).
  2. [Table 1 and Section 4.3] The power-iteration recipe in Eq. (5) is not correct as stated for computing λ2(FFᵀ) (the smallest eigenvalue). With X = FFᵀ − λ1I, the largest-magnitude eigenvalue of X is λ1 − λmin, not λmin. The two unnormalized iterations (p ← Xq, q ← Xp) and the quotient ||q||/||p|| estimate the largest eigenvalue of X, which is λ1 − λmin, so the procedure returns the spectral gap, not λ2(FFᵀ). To obtain λmin one would instead compute λmin = λ1 − λmax(λ1I − FFᵀ). The text says "X ... is FFᵀ − λ1I for λ2(FFᵀ)", which is inaccurate. Moreover, no normalization is performed in Eq. (5) and no error analysis is provided for the two-step approximation. Please correct the algorithmic description and either supply an error bound or empirically validate that the approximation is close to the exact spectral gap on the feature maps used in the experiments.
minor comments (6)
  1. [Abstract] The abstract contains a typo: "there popular person Re-ID benchmarks" should be "three popular person Re-ID benchmarks".
  2. [Contributions bullet] The phrase "conditional number" should be "condition number".
  3. [Section 4.3, last bullet] The text says ABD-Net (XE) "outperforms 'no diversity' counterpart (baseline (XE) + O.F. + O.W.)", but baseline (XE) + O.F. + O.W. includes diversity. The "no diversity" counterpart should be baseline (XE) + PAM + CAM, which has mAP 78.50 and gives the reported 7.40 mAP difference.
  4. [Section 3.2, Eq. (5)] The notation λ(X) for the eigenvalue estimate and the variables p and q are not formally defined; please clarify the update and the quotient, and specify whether q is renormalized between iterations.
  5. [Section 4.2] The hyperparameters βOF and βOW are written with inconsistent subscript formatting in Eq. (6) and the text; please unify the notation.
  6. [Figure 4 caption] The caption says "O.F. is applied after CAM on res conv 2 and after res conv 5 in the Attentive Branch", but the text also mentions O.F. after the reduction layer and after the CAM/PAM outputs; please make the placement description consistent.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: ABD-Net is an empirical architecture evaluated on external Re-ID benchmarks, and its regularizers are hand-designed rather than fitted predictions.

full rationale

The paper's claims are supported by empirical evaluations on externally defined benchmarks (Market-1501, DukeMTMC-Re-ID, MSMT17) using standard metrics (top-1 and mAP), with ablations in Table 1 and comparisons in Tables 2-4. The SVDO regularizer in Eqs. (3)-(5) is a hand-designed objective, not a quantity fitted to the reported results, and no predicted metric is equivalent to a loss term by construction. The attention modules and orthogonality terms are motivated by prior work, including [13] for orthogonality in Re-ID and [39] for soft spectral regularizers; while [39] shares an author with the present paper, that citation is used as background context and the novel SVDO objective is presented as a distinct contribution, so the self-citation is not load-bearing. The known discrepancy between Eq. (3)'s condition-number ratio and Eq. (4)'s eigenvalue difference is a mathematical fidelity concern about the stated mechanism, not a circularity: the input objective is not defined in terms of the output performance, and the reported gains could in principle fail even if the derivation were correct. Thus no step reduces by definition or by self-citation to its own inputs.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claim rests on standard deep-learning assumptions (pretrained backbone, augmentation, loss tuning) and on one unproved mathematical assertion: that Eq. (4) is a faithful, efficiently computable surrogate for Eq. (3). The latter is recorded as an ad hoc axiom. Hyperparameters are hand-chosen and reported without sensitivity analysis.

free parameters (6)
  • beta_O.F. (feature orthogonality weight) = 1e-6
    Chosen by hand; controls strength of feature decorrelation penalty in loss (6); directly affects the claimed diversity benefit.
  • beta_O.W. (weight orthogonality weight) = 1e-3
    Chosen by hand; controls the strength of weight orthogonality and is central to the diversity mechanism.
  • beta_tr (triplet loss weight) = 0.1
    Chosen by hand; balances triplet loss against cross-entropy and orthogonality terms.
  • Triplet margin alpha = 1.2
    Set for hard mining triplet loss; standard hyperparameter.
  • CAM scaling gamma = not reported in main text
    Scales the channel attention residual in Eq. (2); the value is not given in the text, making exact reproduction incomplete.
  • Power iteration steps = 2 (default)
    Number of iterations approximating eigenvalues in Eq. (5); the approximation error is not analyzed.
assumptions (5)
  • domain assumption ImageNet-pretrained ResNet-50 is an appropriate backbone and fine-tunes successfully to Re-ID tasks.
    Section 4.2 uses ImageNet-pretrained ResNet-50 and the two-step transfer algorithm; this is a standard but unproved assumption for this method.
  • domain assumption Channel and position attention modules from DANet transfer to person Re-ID without modification and complement each other.
    Section 3.1 adopts the designs from Fu et al. [40]; the paper provides no proof they generalize, only empirical ablations.
  • domain assumption Reducing feature correlation via orthogonality improves Euclidean-distance retrieval.
    Sections 2.3 and 3.2 rely on Sun et al. [13]; this is the motivating premise of the diversity term.
  • ad hoc to paper The spectral value difference objective in Eq. (4) is a valid surrogate for the condition number objective in Eq. (3).
    The paper converts Eq. (3) to Eq. (4) without derivation; the two are not equivalent (difference versus ratio), and the approximation via two power iterations is unanalyzed.
  • domain assumption Two-step training with random erasing and the chosen hyperparameters is a fair and stable protocol for comparing Re-ID methods.
    Section 4.2; standard in the literature but affects all reported numbers.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ABD-Net: Attentive but Diverse Person Re-Identification." pith.science (2026). https://pith.science/paper/5DAIPN7E

@misc{pith2026190801114,
  author       = {Pith},
  title        = {Pith review of: ABD-Net: Attentive but Diverse Person Re-Identification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5DAIPN7E}},
  note         = {Machine review of arXiv:1908.01114}
}
read the original abstract

Attention mechanism has been shown to be effective for person re-identification (Re-ID). However, the learned attentive feature embeddings which are often not naturally diverse nor uncorrelated, will compromise the retrieval performance based on the Euclidean distance. We advocate that enforcing diversity could greatly complement the power of attention. To this end, we propose an Attentive but Diverse Network (ABD-Net), which seamlessly integrates attention modules and diversity regularization throughout the entire network, to learn features that are representative, robust, and more discriminative. Specifically, we introduce a pair of complementary attention modules, focusing on channel aggregation and position awareness, respectively. Furthermore, a new efficient form of orthogonality constraint is derived to enforce orthogonality on both hidden activations and weights. Through careful ablation studies, we verify that the proposed attentive and diverse terms each contributes to the performance gains of ABD-Net. On three popular benchmarks, ABD-Net consistently outperforms existing state-of-the-art methods.

Figures

Figures reproduced from arXiv: 1908.01114 by the authors.

Figure 1
Figure 1. Visualization of attention maps. (i) Original images; (ii) [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Channel Attention Module (CAM) 3. Attentive but Diverse Network In this section, we first introduce the two attention mod￾ules, followed by the new diversity (orthogonality) regular￾ization. We then wrap them up and describe the overall ar￾chitecture of ABD-Net. 3.1. Attention: Channel-Wise and Position-Wise The goal of attention for Re-ID is to focus on person￾related features while eliminating irrelevant backgroun… view at source ↗
Figure 4
Figure 4. Architecture of ABD-Net: O.W. is applied on all ResNet layers. O.F. is applied after CAM on res [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (4 more)
Figure 6
Figure 6. Figure 6: Visualization of correlation matrix between channels [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 5
Figure 5. Figure 5: Visualization of attention maps from Baseline, Baseline [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 7
Figure 7. Figure 7: Histogram of correlation from Baseline, Baseline + PAM [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 9
Figure 9. Figure 9: Six Re-ID examples of ABD-Net (XE), Baseline + PAM [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

74 extracted references · 58 canonical work pages

  1. [1]

    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 The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2018. 1, 2, 5

    work page 2018

  2. [2]

    Viewpoint invariant pedes- trian recognition with an ensemble of localized features

    Douglas Gray and Hai Tao. Viewpoint invariant pedes- trian recognition with an ensemble of localized features. In European conference on computer vision , pages 262–275. Springer, 2008. 1, 2

    work page 2008

  3. [3]

    Person re-identification by support vec- tor ranking

    Bryan James Prosser, Wei-Shi Zheng, Shaogang Gong, Tao Xiang, and Q Mary. Person re-identification by support vec- tor ranking. In BMVC, volume 2, page 6, 2010. 1, 2

    work page 2010

  4. [4]

    Person re-identification by multi-channel parts-based cnn with improved triplet loss function

    De Cheng, Yihong Gong, Sanping Zhou, Jinjun Wang, and Nanning Zheng. Person re-identification by multi-channel parts-based cnn with improved triplet loss function. In Pro- ceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1335–1344, 2016. 1, 2

    work page 2016

  5. [5]

    Pose-driven deep convolutional model for per- son re-identification

    Chi Su, Jianing Li, Shiliang Zhang, Junliang Xing, Wen Gao, and Qi Tian. Pose-driven deep convolutional model for per- son re-identification. In Computer Vision (ICCV), 2017 IEEE International Conference on, pages 3980–3989. IEEE, 2017. 1, 2, 7

    work page 2017

  6. [6]

    Pose invariant embedding for deep person re-identification

    Liang Zheng, Yujia Huang, Huchuan Lu, and Yi Yang. Pose invariant embedding for deep person re-identification. arXiv preprint arXiv:1701.07732, 2017. 1, 2

    arXiv 2017

  7. [7]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Advances in Neural Information Processing Systems, pages 5998–6008, 2017. 1

    2017

  8. [8]

    Deeply-learned part-aligned representations for person re- identification

    Liming Zhao, Xi Li, Yueting Zhuang, and Jingdong Wang. Deeply-learned part-aligned representations for person re- identification. In ICCV, pages 3239–3248, 2017. 1, 2

    work page 2017

Show all 74 references
  1. [9]

    Deep representation learning with part loss for person re-identification

    Hantao Yao, Shiliang Zhang, Yongdong Zhang, Jintao Li, and Qi Tian. Deep representation learning with part loss for person re-identification. arXiv preprint arXiv:1707.00798 ,

    arXiv

  2. [10]

    Dual attention matching network for context-aware feature sequence based person re-identification

    Jianlou Si, Honggang Zhang, Chun-Guang Li, Jason Kuen, Xiangfei Kong, Alex C Kot, and Gang Wang. Dual attention matching network for context-aware feature sequence based person re-identification. arXiv preprint arXiv:1803.09937 ,

    arXiv

  3. [11]

    Attention-aware compositional network for person re-identification

    Jing Xu, Rui Zhao, Feng Zhu, Huaming Wang, and Wanli Ouyang. Attention-aware compositional network for person re-identification. arXiv preprint arXiv:1805.03344, 2018. 1, 2, 4, 6, 7

    2018 arXiv

  4. [12]

    Harmonious at- tention network for person re-identification

    Wei Li, Xiatian Zhu, and Shaogang Gong. Harmonious at- tention network for person re-identification. In CVPR, vol- ume 1, page 2, 2018. 1, 2

    2018

  5. [13]

    Svdnet for pedestrian retrieval

    Yifan Sun, Liang Zheng, Weijian Deng, and Shengjin Wang. Svdnet for pedestrian retrieval. 2017 IEEE International Conference on Computer Vision (ICCV), Oct 2017. 1, 2, 3, 5, 6, 7

    2017

  6. [14]

    Scalable person re-identification: A benchmark

    Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jing- dong Wang, and Qi Tian. Scalable person re-identification: A benchmark. In The IEEE International Conference on Computer Vision (ICCV), December 2015. 2, 5

    2015

  7. [15]

    Performance measures and a data set for multi-target, multi-camera tracking

    Ergys Ristani, Francesco Solera, Roger Zou, Rita Cucchiara, and Carlo Tomasi. Performance measures and a data set for multi-target, multi-camera tracking. In The European Con- ference on Computer Vision (ECCV) , September 2016. 2, 5

    2016

  8. [16]

    Distance metric learning for large margin nearest neighbor classification

    Kilian Q Weinberger, John Blitzer, and Lawrence K Saul. Distance metric learning for large margin nearest neighbor classification. In Advances in neural information processing systems, pages 1473–1480, 2006. 2

    2006

  9. [17]

    Learning locally-adaptive de- cision functions for person verification

    Zhen Li, Shiyu Chang, Feng Liang, Thomas S Huang, Lian- gliang Cao, and John R Smith. Learning locally-adaptive de- cision functions for person verification. InProceedings of the IEEE Conference on Computer Vision and Pattern Recogni- tion, pages 3610–3617, 2013. 2

    2013

  10. [18]

    Joint learning for attribute- consistent person re-identification

    Sameh Khamis, Cheng-Hao Kuo, Vivek K Singh, Vinay D Shet, and Larry S Davis. Joint learning for attribute- consistent person re-identification. In European Conference on Computer Vision, pages 134–146. Springer, 2014. 2

    2014

  11. [19]

    Large scale metric learning from equivalence constraints

    Martin Koestinger, Martin Hirzer, Paul Wohlhart, Peter M Roth, and Horst Bischof. Large scale metric learning from equivalence constraints. In Computer Vision and Pat- tern Recognition (CVPR), 2012 IEEE Conference on , pages 2288–2295. IEEE, 2012. 2

    2012

  12. [20]

    Locally aligned feature trans- forms across views

    Wei Li and Xiaogang Wang. Locally aligned feature trans- forms across views. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 3594– 3601, 2013. 2

    2013

  13. [21]

    Bicov: a novel im- age representation for person re-identification and face ver- ification

    Bingpeng Ma, Yu Su, and Fr ´ed´eric Jurie. Bicov: a novel im- age representation for person re-identification and face ver- ification. In British Machive Vision Conference, pages 11– pages, 2012. 2

    2012

  14. [22]

    Deep- reid: Deep filter pairing neural network for person re- identification

    Wei Li, Rui Zhao, Tong Xiao, and Xiaogang Wang. Deep- reid: Deep filter pairing neural network for person re- identification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 152–159,

  15. [23]

    Learning mid-level filters for person re-identification

    Rui Zhao, Wanli Ouyang, and Xiaogang Wang. Learning mid-level filters for person re-identification. In Proceed- ings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 144–151, 2014. 2

    2014

  16. [24]

    In de- fense of the triplet loss for person re-identification

    Alexander Hermans, Lucas Beyer, and Bastian Leibe. In de- fense of the triplet loss for person re-identification. arXiv preprint arXiv:1703.07737, 2017. 2

    2017 arXiv

  17. [25]

    Triplet-based deep similarity learning for person re-identification

    Wentong Liao, Michael Ying Yang, Ni Zhan, and Bodo Rosenhahn. Triplet-based deep similarity learning for person re-identification. In Proceedings of the 2017 IEEE Interna- tional Conference on Computer Vision Workshop (ICCVW), pages 385–393, 2017. 2

    2017

  18. [26]

    Per- son re-identification by local maximal occurrence represen- tation and metric learning

    Shengcai Liao, Yang Hu, Xiangyu Zhu, and Stan Z Li. Per- son re-identification by local maximal occurrence represen- tation and metric learning. In Proceedings of the IEEE con- ference on computer vision and pattern recognition , pages 2197–2206, 2015. 2

    2015

  19. [27]

    Domain transfer support vector ranking for person re-identification without target camera label information

    Andy J Ma, Pong C Yuen, and Jiawei Li. Domain transfer support vector ranking for person re-identification without target camera label information. In Proceedings of the IEEE International Conference on Computer Vision, pages 3567– 3574, 2013. 2

    2013

  20. [28]

    Reidentifi- cation by relative distance comparison.IEEE transactions on pattern analysis and machine intelligence , 35(3):653–668,

    Wei-Shi Zheng, Shaogang Gong, and Tao Xiang. Reidentifi- cation by relative distance comparison.IEEE transactions on pattern analysis and machine intelligence , 35(3):653–668,

  21. [29]

    Glad: global-local-alignment descriptor for pedes- trian retrieval

    Longhui Wei, Shiliang Zhang, Hantao Yao, Wen Gao, and Qi Tian. Glad: global-local-alignment descriptor for pedes- trian retrieval. In Proceedings of the 2017 ACM on Multime- dia Conference, pages 420–428. ACM, 2017. 2, 7

    2017

  22. [30]

    Part-based deep hashing for large-scale person re-identification

    Fuqing Zhu, Xiangwei Kong, Liang Zheng, Haiyan Fu, and Qi Tian. Part-based deep hashing for large-scale person re-identification. IEEE Transactions on Image Processing , 26(10):4806–4817, 2017. 2

    2017

  23. [31]

    Part-aligned bilinear representations for person re-identification

    Yumin Suh, Jingdong Wang, Siyu Tang, Tao Mei, and Ky- oung Mu Lee. Part-aligned bilinear representations for person re-identification. arXiv preprint arXiv:1804.07094 ,

    arXiv

  24. [32]

    Spindle net: Person re-identification with human body region guided feature decomposition and fusion

    Haiyu Zhao, Maoqing Tian, Shuyang Sun, Jing Shao, Junjie Yan, Shuai Yi, Xiaogang Wang, and Xiaoou Tang. Spindle net: Person re-identification with human body region guided feature decomposition and fusion. In Proceedings of the IEEE Conference on Computer Vision and Pattern Rec...

    2017

  25. [33]

    Per- son re-identification: Past, present and future

    Liang Zheng, Yi Yang, and Alexander G Hauptmann. Per- son re-identification: Past, present and future. arXiv preprint arXiv:1610.02984, 2016. 2

    2016 arXiv

  26. [34]

    Chris Colbert, and Ga ¨el Varoquaux

    St ´efan van der Walt, S. Chris Colbert, and Ga ¨el Varoquaux. The numpy array: a structure for efficient numerical compu- tation. CoRR, abs/1102.1523, 2011. 2, 3

    2011 arXiv

  27. [35]

    Generalized backpropagation,\’{E} tude de cas: Orthogonality

    Mehrtash Harandi and Basura Fernando. Generalized backpropagation,\’{E} tude de cas: Orthogonality. arXiv preprint arXiv:1611.05927, 2016. 2, 3

    2016 arXiv

  28. [36]

    Optimization on subman- ifolds of convolution kernels in cnns, 2016

    Mete Ozay and Takayuki Okatani. Optimization on subman- ifolds of convolution kernels in cnns, 2016. 2, 3

    2016

  29. [37]

    Orthogonal weight normaliza- tion: Solution to optimization over multiple dependent stiefel manifolds in deep neural networks, 2017

    Lei Huang, Xianglong Liu, Bo Lang, Adams Wei Yu, Yongliang Wang, and Bo Li. Orthogonal weight normaliza- tion: Solution to optimization over multiple dependent stiefel manifolds in deep neural networks, 2017. 2, 3

    2017

  30. [38]

    All you need is beyond a good init: Exploring better solution for training extremely deep convolutional neural networks with orthonormality and modulation

    Di Xie, Jiang Xiong, and Shiliang Pu. All you need is beyond a good init: Exploring better solution for training extremely deep convolutional neural networks with orthonormality and modulation. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jul 2017. 2, 3

    2017

  31. [39]

    Can we gain more from orthogonality regularizations in training deep networks? In Advances in Neural Information Process- ing Systems, pages 4266–4276, 2018

    Nitin Bansal, Xiaohan Chen, and Zhangyang Wang. Can we gain more from orthogonality regularizations in training deep networks? In Advances in Neural Information Process- ing Systems, pages 4266–4276, 2018. 2, 3, 4

    2018

  32. [40]

    Dual attention network for scene segmentation

    Jun Fu, Jing Liu, Haijie Tian, Zhiwei Fang, and Hanqing Lu. Dual attention network for scene segmentation. arXiv preprint arXiv:1809.02983, 2018. 3

    2018 arXiv

  33. [41]

    Learning simple thresholded features with sparse support recovery

    Hongyu Xu, Zhangyang Wang, Haichuan Yang, Ding Liu, and Ji Liu. Learning simple thresholded features with sparse support recovery. IEEE Transactions on Circuits and Sys- tems for Video Technology, 2019. 3

    2019

  34. [42]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 4

    2016

  35. [43]

    Inception-v4, inception-resnet and the impact of residual connections on learning

    Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke, and Alexander A Alemi. Inception-v4, inception-resnet and the impact of residual connections on learning. In Thirty-First AAAI Conference on Artificial Intelligence, 2017. 4

    2017

  36. [44]

    Densely connected convolutional net- works

    Gao Huang, Zhuang Liu, Laurens Van Der Maaten, and Kil- ian Q Weinberger. Densely connected convolutional net- works. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4700–4708, 2017. 4

    2017

  37. [45]

    Support neighbor loss for person re-identification

    Kai Li, Zhengming Ding, Kunpeng Li, Yulun Zhang, and Yun Fu. Support neighbor loss for person re-identification. In 2018 ACM Multimedia Conference on Multimedia Con- ference, pages 1492–1500. ACM, 2018. 4, 6

    2018

  38. [46]

    Person re- identification with hierarchical deep learning feature and ef- ficient xqda metric

    Mingyong Zeng, Chang Tian, and Zemin Wu. Person re- identification with hierarchical deep learning feature and ef- ficient xqda metric. In 2018 ACM Multimedia Conference on Multimedia Conference, pages 1838–1846. ACM, 2018. 4, 6

    2018

  39. [47]

    Learning discriminative features with multiple granularities for person re-identification

    Guanshuo Wang, Yufeng Yuan, Xiong Chen, Jiwei Li, and Xi Zhou. Learning discriminative features with multiple granularities for person re-identification. In 2018 ACM Mul- timedia Conference on Multimedia Conference , pages 274–

    2018

  40. [48]

    Local convolutional neural networks for person re-identification

    Jiwei Yang, Xu Shen, Xinmei Tian, Houqiang Li, Jianqiang Huang, and Xian-Sheng Hua. Local convolutional neural networks for person re-identification. In 2018 ACM Multi- media Conference on Multimedia Conference , pages 1074–

    2018

  41. [49]

    Mask-guided contrastive attention model for person re-identification

    Chunfeng Song, Yan Huang, Wanli Ouyang, and Liang Wang. Mask-guided contrastive attention model for person re-identification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 1179– 1188, 2018. 4, 6

    2018

  42. [50]

    Harmonious at- tention network for person re-identification.2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition , Jun 2018

    Wei Li, Xiatian Zhu, and Shaogang Gong. Harmonious at- tention network for person re-identification.2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition , Jun 2018. 4, 6, 7

    2018

  43. [51]

    Jiawei Liu, Zheng-Jun Zha, Hongtao Xie, Zhiwei Xiong, and Yongdong Zhang. Ca3net. 2018 ACM Multimedia Confer- ence on Multimedia Conference - MM 18, 2018. 4, 6, 7

    2018

  44. [52]

    Random erasing data augmentation

    Zhun Zhong, Liang Zheng, Guoliang Kang, Shaozi Li, and Yi Yang. Random erasing data augmentation. arXiv preprint arXiv:1708.04896, 2017. 5, 6, 7

    2017 arXiv

  45. [53]

    Deep transfer learning for person re-identification

    Mengyue Geng, Yaowei Wang, Tao Xiang, and Yonghong Tian. Deep transfer learning for person re-identification. arXiv preprint arXiv:1611.05244, 2016. 5

    2016 arXiv

  46. [54]

    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 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1318–1327, 2017. 6

    2017

  47. [55]

    Scalable person re-identification: A benchmark

    Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jing- dong Wang, and Qi Tian. Scalable person re-identification: A benchmark. In Computer Vision, IEEE International Con- ference on, 2015. 6, 7

    2015

  48. [56]

    Scalable person re- identification on supervised smoothed manifold

    Song Bai, Xiang Bai, and Qi Tian. Scalable person re- identification on supervised smoothed manifold. In Proceed- ings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 2530–2539, 2017. 6

    2017

  49. [57]

    Features for multi-target multi-camera tracking and re-identification

    Ergys Ristani and Carlo Tomasi. Features for multi-target multi-camera tracking and re-identification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 6036–6046, 2018. 6, 7

    2018

  50. [58]

    Deep spatial feature reconstruction for partial person re- identification: Alignment-free approach

    Lingxiao He, Jian Liang, Haiqing Li, and Zhenan Sun. Deep spatial feature reconstruction for partial person re- identification: Alignment-free approach. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 7073–7082, 2018. 6

    2018

  51. [59]

    Multi-level factorisation net for person re-identification

    Xiaobin Chang, Timothy M Hospedales, and Tao Xiang. Multi-level factorisation net for person re-identification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 2109–2118, 2018. 6, 7

    2018

  52. [60]

    Group consistent similarity learning via deep crf for person re-identification

    Dapeng Chen, Dan Xu, Hongsheng Li, Nicu Sebe, and Xiao- gang Wang. Group consistent similarity learning via deep crf for person re-identification. In Proceedings of the IEEE Con- ference on Computer Vision and Pattern Recognition, pages 8649–8658, 2018. 6

    2018

  53. [61]

    End-to-end deep kronecker-product matching for person re-identification

    Yantao Shen, Tong Xiao, Hongsheng Li, Shuai Yi, and Xi- aogang Wang. End-to-end deep kronecker-product matching for person re-identification. In Proceedings of the IEEE Con- ference on Computer Vision and Pattern Recognition, pages 6886–6895, 2018. 6

    2018

  54. [62]

    Hard-aware point-to-set deep metric for person re-identification

    Rui Yu, Zhiyong Dou, Song Bai, Zhaoxiang Zhang, Yongchao Xu, and Xiang Bai. Hard-aware point-to-set deep metric for person re-identification. In Proceedings of the Eu- ropean Conference on Computer Vision (ECCV), pages 188– 204, 2018. 6

    2018

  55. [63]

    Person re-identification with deep similarity-guided graph neural network

    Yantao Shen, Hongsheng Li, Shuai Yi, Dapeng Chen, and Xiaogang Wang. Person re-identification with deep similarity-guided graph neural network. In Proceedings of the European Conference on Computer Vision (ECCV) , pages 486–504, 2018. 6

    2018

  56. [64]

    Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline)

    Yifan Sun, Liang Zheng, Yi Yang, Qi Tian, and Shengjin Wang. Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline). In Pro- ceedings of the European Conference on Computer Vision (ECCV), pages 480–496, 2018. 6, 7

    2018

  57. [65]

    Mancs: A multi-task attentional network with curriculum sampling for person re-identification

    Cheng Wang, Qian Zhang, Chang Huang, Wenyu Liu, and Xinggang Wang. Mancs: A multi-task attentional network with curriculum sampling for person re-identification. In Proceedings of the European Conference on Computer Vi- sion (ECCV), pages 365–381, 2018. 6, 7

    2018

  58. [66]

    Attribute-aware attention model for fine-grained representa- tion learning

    Kai Han, Jianyuan Guo, Chao Zhang, and Mingjian Zhu. Attribute-aware attention model for fine-grained representa- tion learning. In 2018 ACM Multimedia Conference on Mul- timedia Conference, pages 2040–2048. ACM, 2018. 6

    2018

  59. [67]

    Human seman- tic parsing for person re-identification

    Mahdi M Kalayeh, Emrah Basaran, Muhittin G ¨okmen, Mustafa E Kamasak, and Mubarak Shah. Human seman- tic parsing for person re-identification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recogni- tion, pages 1062–1071, 2018. 6, 7

    2018

  60. [68]

    Kot, and Gang Wang

    Jianlou Si, Honggang Zhang, Chun-Guang Li, Jason Kuen, Xiangfei Kong, Alex C. Kot, and Gang Wang. Dual atten- tion matching network for context-aware feature sequence based person re-identification. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun 2018. 6, 7

    2018

  61. [69]

    Let features decide for themselves: Feature mask network for person re-identification, 2017

    Guodong Ding, Salman Khan, Zhenmin Tang, and Fatih Porikli. Let features decide for themselves: Feature mask network for person re-identification, 2017. 7

    2017

  62. [70]

    Pedestrian alignment network for large-scale person re-identification,

    Zhedong Zheng, Liang Zheng, and Yi Yang. Pedestrian alignment network for large-scale person re-identification,

  63. [71]

    Deep-person: Learning discrim- inative deep features for person re-identification, 2017

    Xiang Bai, Mingkun Yang, Tengteng Huang, Zhiyong Dou, Rui Yu, and Yongchao Xu. Deep-person: Learning discrim- inative deep features for person re-identification, 2017. 7

    2017

  64. [72]

    Re-id done right: towards good practices for person re- identification, 2018

    Jon Almazan, Bojana Gajic, Naila Murray, and Diane Lar- lus. Re-id done right: towards good practices for person re- identification, 2018. 7

    2018

  65. [73]

    Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, and Dhruv Ba- tra

    Ramprasaath R. Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, and Dhruv Ba- tra. Grad-cam: Visual explanations from deep networks via gradient-based localization. 2017 IEEE International Con- ference on Computer Vision (ICCV), Oct 2017. 6

    2017

  66. [74]

    Towards rich feature discov- ery with class activation maps augmentation for person re- identification

    Wenjie Yang, Houjing Huang, Zhang Zhang, Xiaotang Chen, Kaiqi Huang, and Shu Zhang. Towards rich feature discov- ery with class activation maps augmentation for person re- identification. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2019. 6

    2019

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.