Pith. sign in

REVIEW 4 major objections 4 minor 64 references

It Takes Two: Accurate Gait Recognition in the Wild via Cross-granularity Alignment

T0 review · 4 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Pith's one-line read The paper claims that silhouettes and human parsing maps should be fused, not chosen between, and reports Rank-1 accuracy of 80.5% on Gait3D and 88.3% on CCPG.

desk verdict First credible fusion of silhouette and parsing for gait, with a solid Gait3D result; the CCPG headline is undermined by a possible train/test leakage in the parsing-generation pipeline. read the letter →

arxiv 2411.10742 v1 pith:TI6TQRMV submitted 2024-11-16 cs.CV

classification cs.CV
keywords gaitrecognitionsilhouettesequencehumanparsingcross-granularityalignmentin-the-wildcloth-changingappearancerepresentationfusionbiometrics
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 argues that the two strongest appearance representations for gait recognition — binary silhouettes and human parsing maps — are complementary and should be fused rather than chosen between. Silhouettes give stable segmentation under environmental variation but low information entropy; parsing maps carry fine-grained part dynamics and higher entropy but are less reliably segmented. The proposed XGait encodes both sequences with separate backbones, then aligns them globally with the Global Cross-granularity Module and at the part level with the Part Cross-granularity Module under a learnable part division. The paper reports Rank-1 accuracy of 80.5% on Gait3D and 88.3% on CCPG, and its ablation and intersection experiments indicate that both segmentation quality and information content contribute. A reader should care because this reframes in-the-wild gait recognition as a fusion problem with an explicit quality-versus-entropy trade-off.

What carries the argument

The engine of the method is cross-granularity alignment of two appearance streams. The Global Cross-granularity Module takes globally pooled silhouette and parsing features, concatenates them, passes them through two fully connected layers with a sigmoid, and uses the resulting element-wise weights to recombine the two feature maps. The Part Cross-granularity Module horizontally splits the silhouette feature into top-quarter, middle-half, and bottom-quarter bands, while the parsing feature is divided into upper, middle, and lower body regions using parsing categories; a learnable parameter $\gamma_i$ blends each part mask with its complement before three independent Cross-granularity Alignment heads (CA-Upper, CA-Middle, CA-Down) align the part features. After convergence the learned $\gamma$ values are 1.1, 2.0, and 1.4, meaning the model up-weights middle and lower body while nearly ignoring the upper-body region, consistent with heads, hands, and feet being the most occlusion-prone parts. Four Feature Mapping Heads built from Set Pooling and Horizontal Pyramid Mapping compress each stream, and the final descriptor is the channel-wise concatenation, trained with cross-entropy and triplet losses and matched by Euclidean distance.

What would settle it

Inspect the CCPG fine-tuning protocol: if any of the 1,400 labeled images depicts a subject from the 100-person test set, the parsing inputs are identity-tainted. The decisive check is to regenerate CCPG parsing maps with a model trained only on external data or on a cleanly separated holdout, then re-evaluate Rank-1; a drop below 88.3% would show that part of the reported gain came from leakage rather than the architecture.

Watch

Extended reading notes

Core claim

The paper's central claim is that binary silhouette sequences and gait parsing sequences are not competing inputs but complementary ones, and that a network designed to align them across granularities can beat either representation alone on in-the-wild gait recognition. Silhouettes deliver accurate segmentation and stability under environmental variation but low information entropy, while parsing delivers fine-grained body-part dynamics and higher entropy but degrades under occlusion and clothing variation. XGait is presented as the first framework to integrate these two representations: independent ResNet-style encoders produce feature maps for both streams; the Global Cross-granularity Module recalibrates parsing features using global silhouette statistics; and the Part Cross-granularity Module aligns upper, middle, and lower body regions across the streams under a learnable division that down-weights likely-occluded parts. The paper reports Rank-1 accuracy of 80.5% on Gait3D and 88.3% on CCPG, with ablations showing that each module contributes and that fusing the two representations outperforms distance-level or naive feature-level fusion. The Section 5 intersection experiment is the keystone: after forcing silhouette and parsing to share identical contours, parsing still beats silhouette, which the paper reads as evidence that the fusion gain reflects parsing's higher information content rather than simply better segmentation.

Load-bearing premise

The CCPG result assumes that the 1,400 RGB images sampled to fine-tune the parsing model contain no test-subject identity; if test identities leaked into that labeling set, the reported 88.3% Rank-1 would overstate generalization.

Editorial extensions

If this is right

  • On Gait3D, using both representations lifts Rank-1 accuracy to 80.5%, compared with 58.7% for silhouette alone and 71.2% for parsing alone, so the fusion claim is directly testable against either single stream.
  • When segmentation quality is artificially equalized by intersecting silhouette and parsing contours, parsing still gives higher Rank-1 accuracy than silhouette on CCPG (82.4% versus 78.3% for CL-FULL), implying that improving parsing quality is a concrete lever for further gains.
  • Ablations show that the Global and Part modules each add independent improvement, and the learnable division beats fixed or simple division, so adaptive part weighting is doing real work under occlusion and clothing change.
  • Because the paper reports that independent backbones and independent mapping heads are required (sharing collapses Rank-1 from 80.5% to 40.3%), future extensions should keep modality-specific encoders rather than tying parameters.

Reading between the lines

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

  • Not claimed in the paper: the 1,400 CCPG images used to fine-tune the parsing model are not explicitly restricted to the training split, so the CCPG evaluation should be re-run with a parsing model trained strictly on non-test data.
  • A natural extension the paper does not explore is applying the same cross-granularity alignment to other paired representations, such as silhouette plus depth or RGB plus optical flow, where the same quality-versus-entropy trade-off may hold.
  • The learned part weights predict that occluding the lower body should hurt XGait more than occluding the head or upper body; the paper does not run that occlusion-localization experiment, but it follows directly from the reported $\gamma$ values.
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

4 major / 4 minor

Summary. The paper proposes XGait, a dual-branch gait recognition framework that fuses silhouette sequences and human parsing sequences. The method uses two independent ResNet-like encoders followed by a Global Cross-granularity Module (GCM) and a Part Cross-granularity Module (PCM) with a learnable division mechanism, and is trained with cross-entropy and triplet losses. On the Gait3D and CCPG benchmarks, the authors report Rank-1 accuracies of 80.5% and 88.3%, respectively, and present ablations supporting the complementary value of the two representations. The paper is a camera-ready ACM MM 2024 paper posted on arXiv.

Significance. If the reported results are valid, XGait is a meaningful step for in-the-wild gait recognition: it is, to my knowledge, the first work to explicitly combine silhouette and parsing sequences in a single architecture, and the GCM/PCM design is a reasonable way to exploit their complementary strengths. The ablation study in Table 5 and the intersection experiment in Section 5 are conceptually clean and provide evidence for the method's internal logic. The paper also includes useful analysis of when parsing degrades relative to silhouette. However, the validity of the CCPG headline result rests on a data-leakage concern in the parsing generation pipeline, which is load-bearing for one of the two main benchmark claims and for the Section 5 scientific conclusion.

major comments (4)
  1. [Section 4.2, CCPG parsing extraction] The paper states that for CCPG, 'we randomly sample and label 1,400 RGB images from the CCPG dataset' and fine-tune CDGNet on them, without specifying that these images are restricted to the official 100-subject training split. CCPG has a fixed 100/100 subject train/test split, so a random sample from the full dataset will almost certainly include test-identity images. If so, the parsing model used to generate the inputs for the CCPG evaluation has been adapted to test identities, and the reported 88.3% Rank-1 in Table 2 (and the Table 7 analysis) may partly reflect data leakage rather than generalization. The authors must either clarify that the 1,400 images are drawn only from the training split, or re-run the CCPG evaluation with a parser trained exclusively on the training split. This is a critical point for the validity of the headline CCPG result.
  2. [Section 4.3 vs. Table 3, Gait3D baseline discrepancy] Table 1 reports GaitBase (a silhouette-only method) at 64.6% Rank-1 on Gait3D, while Table 3 reports 'Only Sil.' at 58.7% Rank-1 under what appears to be the same input modality. If 'Only Sil.' is intended as a silhouette-only baseline within the XGait framework, it is well below the published GaitBase number, which makes the 22.3% improvement attributed to the full method harder to interpret. The authors should explain the discrepancy: whether the backbone, training settings, or evaluation protocol differ from the published GaitBase result, or whether Table 3's baseline is a different configuration. Without this clarification, the relative improvement claim in Section 4.4 is not fully grounded.
  3. [Section 5, Table 7, CCPG intersection experiment] The central scientific claim of Section 5—that parsing's higher information entropy benefits gait recognition even in cloth-changing scenarios—is based on the CCPG results in Table 7, which use parsing maps generated by the CDGNet fine-tuned on the 1,400 images discussed above. If those images include test-identity data, the conclusion is not independently verifiable because the parser has already seen the test subjects. The intersection experiment should be repeated with a parsing model trained only on the CCPG training split, so that the comparison between silhouette and parsing reflects generalization rather than leakage.
  4. [Supplementary Table 9 and hyperparameter selection] The reduction ratio r in the CA modules was selected using test-set Rank-1/mAP on both Gait3D and CCPG (Supplementary Table 9), and the learnable division parameters gamma_i are reported after convergence on the test evaluation. Selecting hyperparameters or reporting converged parameters based on test-set results can make the reported accuracies optimistic. Although the ablation shows the method is fairly robust to r, the authors should state whether any validation split (e.g., a held-out portion of the training set) was used for model selection, or discuss the potential for optimistic bias in the reported numbers.
minor comments (4)
  1. [Section 3.2] Typographical issue: 'the parsing feature maps fF_p' should be 'the parsing feature maps F_p'.
  2. [Section 4.3, Gait3D evaluation] The text says 'reaching a Rank-1 accuracy of 81%' but Table 1 and the abstract report 80.5%; please make the numbers consistent.
  3. [Section 4.5] Typo: 'In this sectoin' should be 'In this section'.
  4. [Section 4.2, Training details] The triplet loss margin (0.2) and the loss weights alpha=beta=1.0 are given without any sensitivity analysis or reference; a brief justification or ablation would strengthen the reproducibility.

Circularity Check

1 steps flagged · score 4.0 of 10

The CCPG headline result is partially compromised because the parsing generator is fine-tuned on images randomly sampled from the full CCPG dataset, while the core XGait fusion method and Gait3D evaluation remain independent.

  1. other [Section 4.2 (Implementation Details), CCPG parsing generation; Section 4.1 (Datasets), CCPG split description.]
    "For the CCPG dataset, where parsing is unavailable, we utilize CDGNet [21] to extract the parsing information. To ensure basic quality standards, we randomly sample and label 1,400 RGB images from the CCPG dataset. Labeling follows the guidelines outlined in [57]. The CDGNet model is then fine-tuned using the parameters published in [57]. Finally, we utilize the optimized CDGNet model to extract the parsing data from the CCPG dataset."

    CCPG is described in Section 4.1 as having 100 subjects for training and another 100 for testing. The paper does not restrict the 1,400 randomly sampled labeled images to the training split; a random sample drawn from the full dataset therefore almost certainly includes images of test identities. The fine-tuned CDGNet is then used to generate the parsing sequences that are fed into XGait at test time. Consequently, the CCPG test inputs are produced by a model that has already been adapted to the test population, so the reported 88.3% Rank-1 is not an independent measure of XGait's generalization on CCPG. This does not affect the Gait3D result or the fusion ablations, and it is a benchmark-contamination issue rather than an internal mathematical circularity.

full rationale

The central derivation chain of XGait is self-contained and not circular. The GCM and PCM modules are described by explicit equations, and their contributions are supported by controlled ablations (Tables 4-6) rather than being defined in terms of the final accuracy. The Gait3D evaluation uses officially provided silhouette and parsing data, and the 80.5% Rank-1 result is therefore a genuine empirical claim. The paper's self-citations, mainly to the authors' prior ParsingGait work [57], are not load-bearing in a circular way: the complementarity of silhouette and parsing is tested by the paper's own intersection experiment and fusion ablations, not merely asserted through citation. The main concern is the CCPG benchmark. The paper states that CDGNet is fine-tuned on 1,400 randomly sampled images from the CCPG dataset without saying they are restricted to the training split. Given the official 100/100 subject split, random sampling from the full dataset essentially guarantees test-identity exposure, meaning the parsing test inputs are not independent of the test subjects. This partially contaminates one of the two headline claims. I do not count the choice of reduction ratio r=16 from both test sets (Table 9) as circularity, though it is an evaluation-protocol risk. Overall, the score is 4 rather than 6 because the Gait3D result and the core architectural contribution retain independent empirical content; only the CCPG benchmark is partially compromised by the preprocessing fit.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The method rests on empirical assumptions about the two input representations and on training hyperparameters tuned on the evaluation benchmarks. The most load-bearing assumption is that the CCPG parsing maps are generated by a model fine-tuned on data disjoint from the test identities; this is not stated in the paper.

free parameters (4)
  • Reduction ratio r in CA modules = 16
    Chosen by ablation on Gait3D and CCPG (Table 9); it controls the hidden width of the global and part attention modules.
  • Triplet loss margin = 0.2
    Set by hand following common practice in gait recognition (Section 4.2).
  • Loss weights alpha and beta = 1.0 and 1.0
    Set equal by hand in Equation 6; no sensitivity analysis is reported.
  • Learnable division parameters gamma_i = Converge to 1.1, 2.0, 1.4
    Learned during training, but the choice of 'learnable versus fixed division' was selected using the Gait3D test set (Table 6).
assumptions (5)
  • domain assumption Parsing sequences carry higher information entropy than silhouettes and improve gait discrimination when segmentation is accurate.
    Motivates the PCM and the whole approach; asserted from Zheng et al. [57] and used in Sections 1 and 3.4.
  • domain assumption Silhouette segmentation is more accurate and robust than parsing segmentation in complex scenes.
    Stated in Section 1 and revisited in Section 5; drives the GCM design where silhouettes correct parsing features.
  • domain assumption The human body can be divided into upper, middle, and lower parts with distinct walking dynamics, and horizontal splitting aligns these parts across the two representations.
    Used in the PCM (Section 3.4), citing GaitPart [9] and MMGaitFormer [5].
  • ad hoc to paper CDGNet fine-tuned on 1,400 labeled CCPG images produces parsing maps of sufficient quality for evaluation.
    Section 4.2; the paper does not verify that the 1,400 images are disjoint from the CCPG test set, so the parsing generator may be adapted to test data.
  • standard math Training with cross-entropy and triplet losses on the Gait3D and CCPG training splits yields a generalizable embedding.
    Standard supervised metric-learning assumption; not proven, but routine in the field.

how reviews work

0 comments
Cite this review

Pith. "Pith review of It Takes Two: Accurate Gait Recognition in the Wild via Cross-granularity Alignment." pith.science (2026). https://pith.science/paper/TI6TQRMV

@misc{pith2026241110742,
  author       = {Pith},
  title        = {Pith review of: It Takes Two: Accurate Gait Recognition in the Wild via Cross-granularity Alignment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TI6TQRMV}},
  note         = {Machine review of arXiv:2411.10742}
}
read the original abstract

Existing studies for gait recognition primarily utilized sequences of either binary silhouette or human parsing to encode the shapes and dynamics of persons during walking. Silhouettes exhibit accurate segmentation quality and robustness to environmental variations, but their low information entropy may result in sub-optimal performance. In contrast, human parsing provides fine-grained part segmentation with higher information entropy, but the segmentation quality may deteriorate due to the complex environments. To discover the advantages of silhouette and parsing and overcome their limitations, this paper proposes a novel cross-granularity alignment gait recognition method, named XGait, to unleash the power of gait representations of different granularity. To achieve this goal, the XGait first contains two branches of backbone encoders to map the silhouette sequences and the parsing sequences into two latent spaces, respectively. Moreover, to explore the complementary knowledge across the features of two representations, we design the Global Cross-granularity Module (GCM) and the Part Cross-granularity Module (PCM) after the two encoders. In particular, the GCM aims to enhance the quality of parsing features by leveraging global features from silhouettes, while the PCM aligns the dynamics of human parts between silhouette and parsing features using the high information entropy in parsing sequences. In addition, to effectively guide the alignment of two representations with different granularity at the part level, an elaborate-designed learnable division mechanism is proposed for the parsing features. Comprehensive experiments on two large-scale gait datasets not only show the superior performance of XGait with the Rank-1 accuracy of 80.5% on Gait3D and 88.3% CCPG but also reflect the robustness of the learned features even under challenging conditions like occlusions and cloth changes.

Figures

Figures reproduced from arXiv: 2411.10742 by the authors.

Figure 1
Figure 1. Comparisons of different gait recognition methods, [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The architecture of our XGait. In the preprocessing stage, the silhouette sequence [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 4
Figure 4. The pipeline of the Part Cross-granularity Module [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figures from the paper (4 more)
Figure 6
Figure 6. Figure 6: The illustration of the silhouette and parsing in [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: The visualization of feature distributions. (a) XGait [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 8
Figure 8. Figure 8: Exemplar results of (a) XGait (Only Silhouette), (b) XGait (Only Parsing), and (c) XGait (Ours). This case shows that [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: Exemplar results of (a) XGait (Only Silhouette), (b) XGait (Only Parsing), and (c) XGait (Ours). This case shows that [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 61 canonical work pages

  1. [1]

    Weizhi An, Shiqi Yu, Yasushi Makihara, Xinhui Wu, Chi Xu, Yang Yu, Rijun Liao, and Yasushi Yagi. 2020. Performance Evaluation of Model-Based Gait on Multi-View Very Large Population Database With Pose Sequences. IEEE TBBIS 2, 4 (2020), 421–430

  2. [2]

    Gunawan Ariyanto and Mark S. Nixon. 2011. Model-based 3D gait biometrics. In IJCB. 1–7

  3. [3]

    Zhe Cao, Gines Hidalgo, Tomas Simon, Shih-En Wei, and Yaser Sheikh. 2021. OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields. IEEE TPAMI 43, 1 (2021), 172–186

  4. [4]

    Hanqing Chao, Yiwei He, Junping Zhang, and Jianfeng Feng. 2019. GaitSet: Regarding Gait as a Set for Cross-View Gait Recognition. In AAAI. 8126–8133

  5. [5]

    Yufeng Cui and Yimei Kang. 2023. Multi-modal Gait Recognition via Effective Spatial-Temporal Feature Fusion. In CVPR. 17949–17957

  6. [6]

    Huanzhang Dou, Pengyi Zhang, Wei Su, Yunlong Yu, Yining Lin, and Xi Li. 2023. GaitGCI: Generative Counterfactual Intervention for Gait Recognition. In CVPR. 5578–5588

  7. [7]

    Chao Fan, Junhao Liang, Chuanfu Shen, Saihui Hou, Yongzhen Huang, and Shiqi Yu. 2023. OpenGait: Revisiting Gait Recognition Towards Better Practicality. In CVPR. 9707–9716

  8. [8]

    Chao Fan, Jingzhe Ma, Dongyang Jin, Chuanfu Shen, and Shiqi Yu. 2024. Skele- tonGait: Gait Recognition Using Skeleton Maps. In AAAI. 1662–1669

Show all 64 references
  1. [9]

    Chao Fan, Yunjie Peng, Chunshui Cao, Xu Liu, Saihui Hou, Jiannan Chi, Yongzhen Huang, Qing Li, and Zhiqiang He. 2020. GaitPart: Temporal Part-Based Model for Gait Recognition. In CVPR. 14213–14221

  2. [10]

    Yang Fu, Shibei Meng, Saihui Hou, Xuecai Hu, and Yongzhen Huang. 2023. GPGait: Generalized Pose-based Gait Recognition. In ICCV. 19595–19604

  3. [11]

    Ju Han and Bir Bhanu. 2006. Individual Recognition Using Gait Energy Image. IEEE TPAMI 28, 2 (2006), 316–322

  4. [12]

    Altab Hossain, Yasushi Makihara, Junqiu Wang, and Yasushi Yagi

    Md. Altab Hossain, Yasushi Makihara, Junqiu Wang, and Yasushi Yagi. 2010. Clothing-invariant gait identification using part-based clothing categorization and adaptive weight control. PR 43, 6 (2010), 2281–2291

  5. [13]

    Saihui Hou, Chunshui Cao, Xu Liu, and Yongzhen Huang. 2020. Gait Lateral Network: Learning Discriminative and Compact Representations for Gait Recog- nition. In ECCV. 382–398

  6. [14]

    Xiaohu Huang, Duowang Zhu, Hao Wang, Xinggang Wang, Bo Yang, Botao He, Wenyu Liu, and Bin Feng. 2021. Context-Sensitive Temporal Feature Learning for Gait Recognition. In ICCV. 12909–12918

  7. [15]

    Haruyuki Iwama, Mayu Okumura, Yasushi Makihara, and Yasushi Yagi. 2012. The OU-ISIR Gait Database Comprising the Large Population Dataset and Per- formance Evaluation of Gait Recognition. IEEE TIFS 7, 5 (2012), 1511–1521

  8. [16]

    Weijia Li, Saihui Hou, Chunjie Zhang, Chunshui Cao, Xu Liu, Yongzhen Huang, and Yao Zhao. 2023. An In-Depth Exploration of Person Re-Identification and Gait Recognition in Cloth-Changing Conditions. In CVPR. 13824–13833

  9. [17]

    Zeyi Li, Pan Wang, Zixuan Wang, and De-chuan Zhan. 2024. Flowgananomaly: Flow-based anomaly network intrusion detection with adversarial learning. CJE 33, 1 (2024), 58–71

  10. [18]

    Rijun Liao, Shiqi Yu, Weizhi An, and Yongzhen Huang. 2020. A model-based gait recognition method with body pose and human prior knowledge. PR 98 (2020)

  11. [19]

    Beibei Lin, Shunli Zhang, and Feng Bao. 2020. Gait Recognition with Multiple- Temporal-Scale 3D Convolutional Neural Network. In ACM MM. 3054–3062

  12. [20]

    Beibei Lin, Shunli Zhang, and Xin Yu. 2021. Gait Recognition via Effective Global-Local Feature Representation and Local Temporal Aggregation. In ICCV. 14648–14656

  13. [21]

    Kunliang Liu, Ouk Choi, Jianming Wang, and Wonjun Hwang. 2022. CDGNet: Class Distribution Guided Network for Human Parsing. In CVPR. 4473–4482

  14. [22]

    Xinchen Liu, Wu Liu, Jinkai Zheng, Chenggang Yan, and Tao Mei. 2020. Be- yond the Parts: Learning Multi-view Cross-part Correlation for Vehicle Re- identification. In ACM MM. 907–915

  15. [23]

    Kang Ma, Ying Fu, Dezhi Zheng, Chunshui Cao, Xuecai Hu, and Yongzhen Huang

  16. [24]

    Yasushi Makihara, Hidetoshi Mannami, and Yasushi Yagi. 2010. Gait Analysis of Gender and Age Using a Large-Scale Multi-view Gait Database. In ACCV. 440–451

  17. [25]

    Niyogi and Edward H

    Sourabh A. Niyogi and Edward H. Adelson. 1994. Analyzing and recognizing walking figures in XYT. In CVPR. 469–474

  18. [26]

    Kohei Shiraga, Yasushi Makihara, Daigo Muramatsu, Tomio Echigo, and Yasushi Yagi. 2016. GEINet: View-invariant gait recognition using a convolutional neural network. In ICB. 1–8

  19. [27]

    Chunfeng Song, Yongzhen Huang, Weining Wang, and Liang Wang. 2023. CASIA- E: A Large Comprehensive Dataset for Gait Recognition.IEEE TPAMI 45, 3 (2023), 2801–2815

  20. [28]

    Noriko Takemura, Yasushi Makihara, Daigo Muramatsu, Tomio Echigo, and Yasushi Yagi. 2018. Multi-view large population gait dataset and its performance evaluation for cross-view gait recognition. IPSJ TCV A10 (2018), 4

  21. [29]

    Daoliang Tan, Kaiqi Huang, Shiqi Yu, and Tieniu Tan. 2006. Efficient Night Gait Recognition Based on Template Matching. In ICPR. 1000–1003

  22. [30]

    Torben Teepe, Johannes Gilg, Fabian Herzog, Stefan Hörmann, and Gerhard Rigoll

  23. [31]

    Torben Teepe, Ali Khan, Johannes Gilg, Fabian Herzog, Stefan Hörmann, and Gerhard Rigoll. 2021. Gaitgraph: Graph Convolutional Network for Skeleton- Based Gait Recognition. In ICIP. 2314–2318

  24. [32]

    Ye Tian, Ying Fu, and Jun Zhang. 2023. Transformer-based under-sampled single- pixel imaging. CJE 32, 5 (2023), 1151–1159

  25. [33]

    Akira Tsuji, Yasushi Makihara, and Yasushi Yagi. 2010. Silhouette transformation based on walking speed for gait identification. In CVPR. 717–722

  26. [34]

    Zasim Uddin, Trung Ngo Thanh, Yasushi Makihara, Noriko Takemura, Xiang Li, Daigo Muramatsu, and Yasushi Yagi

    Md. Zasim Uddin, Trung Ngo Thanh, Yasushi Makihara, Noriko Takemura, Xiang Li, Daigo Muramatsu, and Yasushi Yagi. 2018. The OU-ISIR Large Population Gait Database with real-life carried object and its performance evaluation. IPSJ TCV A10 (2018), 5

  27. [35]

    Raquel Urtasun and Pascal Fua. 2004. 3D Tracking for Gait Characterization and Recognition. In FGR. 17–22

  28. [36]

    Laurens Van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-SNE. JMLR 9, 11 (2008)

  29. [37]

    Changsheng Wan, Li Wang, and Vir V. Phoha. 2019. A Survey on Gait Recognition. ACM CSUR 51, 5 (2019), 89:1–89:35

  30. [38]

    Chen Wang, Junping Zhang, Jian Pu, Xiaoru Yuan, and Liang Wang. 2010. Chrono- Gait Image: A Novel Temporal Template for Gait Recognition. InECCV, Vol. 6311. 257–270

  31. [39]

    Jingdong Wang, Ke Sun, Tianheng Cheng, Borui Jiang, Chaorui Deng, Yang Zhao, Dong Liu, Yadong Mu, Mingkui Tan, Xinggang Wang, Wenyu Liu, and Bin Xiao

  32. [40]

    Lei Wang, Bo Liu, Fangfang Liang, and Bincheng Wang. 2023. Hierarchical Spatio- Temporal Representation Learning for Gait Recognition. In ICCV. 19639–19649

  33. [41]

    Liang Wang, Tieniu Tan, Huazhong Ning, and Weiming Hu. 2003. Silhouette Analysis-Based Gait Recognition for Human Identification. IEEE TPAMI 25, 12 (2003), 1505–1518

  34. [42]

    Ming Wang, Xianda Guo, Beibei Lin, Tian Yang, Zheng Zhu, Lincheng Li, Shunli Zhang, and Xin Yu. 2023. DyGait: Exploiting Dynamic Representations for High-performance Gait Recognition. In ICCV. 13424–13433

  35. [43]

    Zifeng Wu, Yongzhen Huang, Liang Wang, Xiaogang Wang, and Tieniu Tan

  36. [44]

    Chi Xu, Yasushi Makihara, Gakuto Ogi, Xiang Li, Yasushi Yagi, and Jianfeng Lu

  37. [45]

    Nixon, and John N

    Chew-Yean Yam, Mark S. Nixon, and John N. Carter. 2004. Automated person recognition by walking and running via model-based approaches. PR 37, 5 (2004), 1057–1072

  38. [46]

    Koichiro Yamauchi, Bir Bhanu, and Hideo Saito. 2009. Recognition of walking humans in 3D: Initial results. In CVPRW. 45–52

  39. [47]

    Quanwei Yang, Xinchen Liu, Wu Liu, Hongtao Xie, Xiaoyan Gu, Lingyun Yu, and Yongdong Zhang. 2022. REMOT: A Region-to-Whole Framework for Realistic Human Motion Transfer. In ACM MM. 1128–1137

  40. [48]

    Mang Ye, Jianbing Shen, Gaojie Lin, Tao Xiang, Ling Shao, and Steven C. H. Hoi

  41. [49]

    IPSJ TCV A9 (2017), 24

    The OU-ISIR Gait Database comprising the Large Population Dataset with Age and performance evaluation of age estimation. IPSJ TCV A9 (2017), 24

  42. [50]

    Shiqi Yu, Daoliang Tan, and Tieniu Tan. 2006. A Framework for Evaluating the Effect of View Angle, Clothing and Carrying Condition on Gait Recognition. In ICPR. 441–444

  43. [51]

    Cun Zhang, Xing-Peng Chen, Guo-Qiang Han, and Xiang-Jie Liu. 2023. Spatial transformer network on skeleton-based gait recognition. Expert-Systems 40, 6 (2023)

  44. [52]

    Shaoxiong Zhang, Yunhong Wang, and Annan Li. 2021. Cross-View Gait Recog- nition With Deep Universal Linear Embeddings. In CVPR. 9095–9104

  45. [53]

    Tao Zhang, Ying Fu, and Jun Zhang. 2024. Deep Guided Attention Network for Joint Denoising and Demosaicing in Real Image. CJE 33, 1 (2024), 303–312

  46. [54]

    Ziyuan Zhang, Luan Tran, Xi Yin, Yousef Atoum, Xiaoming Liu, Jian Wan, and Nanxin Wang. 2019. Gait Recognition via Disentangled Representation Learning. In CVPR. 4710–4719

  47. [55]

    Zhaoda Ye, Xiangteng He, and Yuxin Peng. 2022. Unsupervised Cross-Media Hashing Learning via Knowledge Graph. CJE 31, 6 (2022), 1081–1091

  48. [56]

    Jinkai Zheng, Xinchen Liu, Wu Liu, Lingxiao He, Chenggang Yan, and Tao Mei. 2022. Gait Recognition in the Wild with Dense 3D Representations and A Benchmark. In CVPR. 20228–20237

  49. [57]

    Jinkai Zheng, Xinchen Liu, Shuai Wang, Lihao Wang, Chenggang Yan, and Wu Liu. 2023. Parsing is All You Need for Accurate Gait Recognition in the Wild. In ACM MM. 116–124

  50. [58]

    Zheng Zhu, Xianda Guo, Tian Yang, Junjie Huang, Jiankang Deng, Guan Huang, Dalong Du, Jiwen Lu, and Jie Zhou. 2021. Gait Recognition in the Wild: A Benchmark. In ICCV. 14789–14799. MM ’24, October 28–November 1, 2024, Melbourne, VIC, Australia. Jinkai Zheng et al. 3982 (a) (b)...

  51. [61]

    Jinkai Zheng, Xinchen Liu, Xiaoyan Gu, Yaoqi Sun, Chuang Gan, Jiyong Zhang, Wu Liu, and Chenggang Yan. 2022. Gait Recognition in the Wild with Multi-hop Temporal Switch. In ACM MM. 6136–6145

  52. [2017]

    IEEE TPAMI 39, 2 (2017), 209–226

    A Comprehensive Study on Cross-View Gait Based Human Identification with Deep CNNs. IEEE TPAMI 39, 2 (2017), 209–226

  53. [2020]

    CoRR abs/2001.04193 (2020)

    Deep Learning for Person Re-identification: A Survey and Outlook. CoRR abs/2001.04193 (2020)

  54. [2021]

    IEEE TPAMI 43, 10 (2021), 3349–3364

    Deep High-Resolution Representation Learning for Visual Recognition. IEEE TPAMI 43, 10 (2021), 3349–3364. It Takes Two: Accurate Gait Recognition in the Wild via Cross-granularity Alignment MM ’24, October 28–November 1, 2024, Melbourne, VIC, Australia. 0453 (a) (b) (c) Figure...

  55. [2022]

    In CVPRW

    Towards a Deeper Understanding of Skeleton-based Gait Recognition. In CVPRW. 1568–1576

  56. [2023]

    Dynamic Aggregated Network for Gait Recognition. InCVPR. 22076–22085

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.