Pith. sign in

REVIEW 3 major objections 5 minor 61 references

Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read A training-only plug-in module with under 0.01M parameters substantially improves skeleton-based recognition of ambiguous actions by synchronizing spatial and temporal feature extraction.

desk verdict F-RL is constant and Eq. (9) cancels the cross-dimensional features, so the paper's two named mechanisms are invalidated by its own equations; the empirical study is thorough but unattributable as written. read the letter →

arxiv 2412.14833 v2 pith:DZFQ5SPL submitted 2024-12-19 cs.CV

Hao Huang , Yujie Lin , Siyu Chen , Haiyang Liu This is my paper · ORCID
classification cs.CV
keywords skeleton-basedactionrecognitionambiguousgraphconvolutionalnetworksspatial-temporalfeatureextractionaggregationplug-and-playmoduleredundancylossattentionmechanism
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 proposes SF-Head, a training-only plug-in module inserted between the GCN and TCN layers of skeleton-based action recognition networks. Its goal is to improve recognition of ambiguous actions—similar motions such as 'waving' versus 'saluting' or 'reading' versus 'writing'—which existing serial GCN-TCN pipelines handle poorly. The module synchronizes spatial and temporal feature extraction (SSTE), balances those features with a Feature Redundancy Loss, and adaptively aggregates channel, temporal, and spatial features (AC-FA) under a Feature Consistency Loss. The authors claim this yields consistent accuracy gains on four benchmarks while adding fewer than 0.01M parameters and no inference-time cost.

What carries the argument

The key machinery is a pair of modules inside SF-Head. SSTE divides the input into channel groups and parallel branches to produce temporal and spatial feature maps via pooling, group normalization, and sigmoid gating; the Feature Redundancy Loss (F-RL) is meant to balance these two feature types. AC-FA applies adaptive attention weights in the channel (ACDA), temporal (ATDA), and spatial (ASDA) dimensions, then aggregates them with the original feature in a cross-dimensional fusion step (C-FA) governed by Eqs. (8) and (9). The Feature Consistency Loss (F-CL) keeps the aggregated feature close to the original spatial-temporal features, with a soft-margin penalty and a compensation term. The paper's argument rests on these equations actually altering the feature dynamics: F-RL uses a modified cosine distance (Eq. 10) inside a contrastive-style objective (Eq. 11), and F-CL combines a soft-margin log term (Eq. 12) with a data-dependent compensation (Eq. 13).

What would settle it

Compute the gradient of the Feature Redundancy Loss (Eq. 11) with respect to the temporal and spatial features $f_t$ and $f_s$ on a real batch from NTU RGB+D 120; if the gradient is zero, the loss cannot be responsible for any observed accuracy gain. Similarly, verify whether Eq. (9)'s indicator term actually replaces the weighted product with a reshaped version; if the aggregation reduces to a weighted sum of unchanged features, the proposed AC-FA mechanism is not what drives the results.

Watch

Extended reading notes

Core claim

The central claim is that ambiguous skeleton actions can be discriminated much better by synchronizing spatial and temporal feature extraction instead of treating GCN and TCN outputs serially, and by adaptively weighting channel, temporal, and spatial feature maps before aggregation. The paper states this explicitly: SF-Head, used only during training and inserted between GCN and TCN blocks, with SSTE plus F-RL and AC-FA plus F-CL, improves accuracy on NTU RGB+D 60, NTU RGB+D 120, NW-UCLA, and PKU-MMD I. Across seven GCN backbones on NTU RGB+D 120, the average gain is about 0.8%, with larger gains on ambiguous action groups such as 'writing, typing, playing with phone, reading' (up to 5.21% for the 'writing' class). The authors present ablation studies and t-SNE visualizations to attribute these gains to the module's components.

Load-bearing premise

The claimed gains depend on the proposed losses and aggregation equations actually altering the feature representations as described: the redundancy loss must have a non-zero gradient that rebalances spatial and temporal features, and the cross-dimensional aggregation must genuinely transform features rather than leaving them effectively unchanged.

Editorial extensions

If this is right

  • Integrating SF-Head into any GCN-based backbone yields consistent accuracy gains on NTU RGB+D 120 X-Sub and X-Set, averaging about 0.8%, with the largest gains on ambiguous action groups.
  • Because SF-Head is used only during training, deployed models incur zero additional inference cost, making the module practical for real-world applications.
  • The module adds fewer than 0.01M parameters, so even large-scale systems can adopt it without meaningful computational overhead.
  • The synchronization and cross-dimensional aggregation improve both local-detail preservation and global context, which the paper argues is the key to separating visually similar actions.

Reading between the lines

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

  • The synchronization idea could transfer beyond skeleton GCN-TCN pipelines, for example to transformer-based or multi-modal action recognition models, although the paper does not test such settings.
  • The consistent gains across backbones with different base accuracies suggest the module may address a general limitation of serial spatial-temporal extraction rather than overfitting to one architecture.
  • A testable extension is applying SF-Head to RGB-skeleton fused models or to fine-grained video classification tasks where classes are distinguished by small temporal differences.
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 introduces SF-Head, a plug-and-play module for skeleton-based action recognition that is inserted between GCN and TCN blocks. SF-Head contains a Synchronized Spatial-Temporal Extraction (SSTE) branch, an Adaptive Cross-dimensional Feature Aggregation (AC-FA) branch, and two auxiliary losses: a Feature Redundancy Loss (F-RL) and a Feature Consistency Loss (F-CL). The authors report accuracy gains on NTU RGB+D 60/120, NW-UCLA, and PKU-MMD I, and demonstrate integration with seven GCN backbones. The claims are that SF-Head has fewer than 0.01M parameters, is training-only, and improves recognition of ambiguous actions by balancing spatial/temporal features and combining global context with local details.

Significance. The practical ingredients of the paper—a lightweight, training-only adapter with consistent gains across backbones and datasets—are potentially useful to the skeleton-based action recognition community. The experimental coverage is broad, including four benchmarks, seven backbone integrations, and explicit per-action analysis for ambiguous classes. However, the central loss term F-RL, as defined in Eq. (11), is a constant with zero gradient, so the mechanism the paper claims for balancing spatial and temporal features is not implemented by the printed equations. This is a load-bearing defect: the ablation and hyperparameter results in Tables 1–3 cannot be attributed to F-RL, and the main methodological claim is unsupported. The apparent cancellation in Eq. (9) is not, in itself, an error if y_t and y_s denote the reshaped attended features, but the notation is confusing and should be clarified.

major comments (3)
  1. [Section 3.3, Eq. (11)] F-RL is constant with respect to all trainable parameters. In Eq. (11), the denominator sum over j=1..N has summand exp(m·d(f_i^s,f_i^t)/τ), which does not depend on j; hence the sum equals N·exp(m·d_i/τ) and L_red = -1/(2N) Σ_i log(1/N) = (1/2) log N exactly. Its gradient with respect to f_i^s and f_i^t vanishes identically, so F-RL cannot balance spatial and temporal features during training. The statements in Sec. 3.3 and the ablation rows involving F-RL (Table 2, rows 4, 6, 8; Fig. 5, right panel) therefore cannot be caused by F-RL. The hyper-parameter study in Table 1 labels λ_red as an F-RL hyper-parameter, but rows vary λ_con and λ_red simultaneously, and with the printed L_red any effect of λ_red is impossible. The loss must be redefined with a nontrivial partition function (e.g., negative pairs d(f_i^s, f_j^t) with j≠i) and the experiments repeated under the corrected loss.
  2. [Section 4.3, Table 1] The hyperparameter selection procedure is conducted directly on the NTU RGB+D 120 X-sub benchmark, which Sec. 4.1 defines as the test protocol. No training/validation split or model selection strategy is described in Sec. 4.2 or Sec. 4.3. Tuning m, γ, λ_con, λ_red on the test benchmark overfits the reported accuracy and undermines the claim of robustness; a held-out validation set or nested cross-validation must be used before final test evaluation.
  3. [Section 4.3, Tables 2–3] The ablation studies report single-run accuracy differences, several of which are very small: SSTE adds 0.09% (Table 2, row 2) and ASDA adds 0.10% (Table 3, row 4). No repeated runs, standard deviations, or significance tests are provided, and the code is not included in the submission. These small deltas are within typical run-to-run variation for this benchmark and do not establish the individual contribution of the sub-modules claimed in the text.
minor comments (5)
  1. [Section 3.2, Eq. (9)] The notation y_i + I_{t,s}(i)(y_i - y_i) is confusing because I_{t,s}(i)(y_i - y_i) is identically zero for the indicated set; the authors should define y_t and y_s explicitly (e.g., y_t = reshape(ω_t ⊙ f̂_t), y_s = reshape(ω_s ⊙ f̂_s)) and write f_a = Σ_i η_i y_i. As printed, this notation invites the misreading that the transformed features cancel, even though Eq. (8) preserves them.
  2. [Section 3.3, Eq. (10)] The text states d ∈ [0,1], but from the definition d = 1 - exp(-α(cos+1)) one only has d ∈ [0, 1 - e^{-2α}] for finite α; the bound on L_con in Sec. 3.3 should be stated with this actual range.
  3. [Section 3.2, Eq. (1)] There is a notation mismatch between the text, which defines AP as averaging pooling, and the equation, which uses GP in the temporal branch; the formula for f_s also has an unbalanced parenthesis.
  4. [Section 4.5, Fig. 3] The statement that accuracy increments are 'significantly higher than the enhancement of 1.7% (from 88.9% to 90.6%, presented in SOTA)' is unclear; the source of the 1.7% figure and the exact comparison protocol should be specified.
  5. [Section 3.3, Eq. (13)] The explanation that 'when d converges to 0 or 1, φ_i reaches the minimum value 0' is misleading; under the piecewise definition φ_i is set to zero in those regions, while for d inside the active region φ_i is negative, so the description of the penalty's behavior should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: SF-Head is an empirical plug-in validated on public benchmarks; no step reduces its claims to its own inputs.

full rationale

The paper's contribution is an empirical, training-only module evaluated on standard public benchmarks (NTU RGB+D 60/120, NW-UCLA, PKU-MMD I) with standard train/test splits. There is no derived first-principles prediction that is equivalent by construction to a fitted input: no parameter is fitted to a subset and then used to 'predict' that same subset, no uniqueness theorem is invoked, and no load-bearing premise is justified only by a self-citation. The module builds on cited prior work (e.g., triplet attention [31], feature calibration [44], hierarchical decomposition [18]) as acknowledged building blocks rather than as an authority that forces the paper's conclusion. The most notable internal issues are non-circular soundness defects: Eq. (11) defines L_red with a softmax denominator whose summand is independent of j, making the loss a constant with zero gradient, and Eq. (9) algebraically cancels the cross-dimensional correction term introduced in Eq. (8). These undermine the attribution of the reported gains to the stated F-RL and AC-FA mechanisms, but they are correctness and reproducibility concerns, not cases of the derivation reducing to its own inputs or of a fitted quantity being renamed as a prediction. The central empirical claims are self-contained against external benchmarks, so the appropriate circularity finding is no significant circularity.

Assumptions & free parameters 9 free parameters · 4 assumptions · 0 invented entities

The central claim relies on the correctness of the backbone and benchmark protocols, plus the described training dynamics. The loss equations as written are defective, and several hyperparameters are either tuned on the test set or left unspecified.

free parameters (9)
  • margin m for F-CL = 0.4
    Tuned on NTU-RGB+D 120 X-sub (Table 1) and used in F-CL; no separate validation set is reported.
  • gamma for F-CL compensation = 2
    Tuned on NTU-RGB+D 120 X-sub (Table 1).
  • lambda_con = 0.2
    Tuned on NTU-RGB+D 120 X-sub (Table 1).
  • lambda_red = 0.1
    Tuned on NTU-RGB+D 120 X-sub (Table 1).
  • alpha in F-RL distance = unspecified
    Eq. (10) introduces alpha but the implementation details do not give its value.
  • temperature tau in F-RL = unspecified
    Eq. (11) uses tau but no value is reported.
  • scale m in F-RL = unspecified
    Eq. (11) uses m and tau, but implementation details only list m for F-CL, not for F-RL.
  • eta_i aggregation weights = unspecified or learned
    Eq. (8) calls eta_i hyper-parameters but no values or learning rule are given.
  • channel group count g = unspecified
    SSTE splits features into g groups along the channel dimension; g is never specified.
assumptions (4)
  • domain assumption Public benchmark datasets and evaluation protocols (NTU RGB+D 60/120, NW-UCLA, PKU-MMD I) are used exactly as in prior work.
    Results depend on standard splits and preprocessing; no preprocessing details are supplied beyond the datasets named.
  • domain assumption The backbone CTR-GCN as implemented reproduces the cited baseline accuracies.
    Ablations compare against CTR-GCN at 84.72 percent on NTU-120 X-sub; the module's gain is relative to this baseline.
  • standard math The feature redundancy and consistency losses can be optimized jointly with cross-entropy as specified.
    This assumes the losses have nonzero gradients; Eq. (11) as written does not.
  • ad hoc to paper The module can be inserted into any GCN-based architecture without retuning the backbone.
    Claimed in Sec. 4.4 and tested on seven backbones with fixed settings, but the paper does not analyze interaction effects.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition." pith.science (2026). https://pith.science/paper/DZFQ5SPL

@misc{pith2026241214833,
  author       = {Pith},
  title        = {Pith review of: Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DZFQ5SPL}},
  note         = {Machine review of arXiv:2412.14833}
}
read the original abstract

Skeleton-based action recognition using GCNs has achieved remarkable performance, but recognizing ambiguous actions, such as "waving" and "saluting", remains a significant challenge. Existing methods typically rely on a serial combination of GCNs and TCNs, where spatial and temporal features are extracted independently, leading to an unbalanced spatial-temporal information, which hinders accurate action recognition. Moreover, existing methods for ambiguous actions often overemphasize local details, resulting in the loss of crucial global context, which further complicates the task of differentiating ambiguous actions. To address these challenges, we propose a lightweight plug-and-play module called SF-Head, inserted between GCN and TCN layers. SF-Head first conducts SSTE with a Feature Redundancy Loss (F-RL), ensuring a balanced interaction. It then performs AC-FA, with a Feature Consistency Loss (F-CL), which aligns the aggregated feature with their original spatial-temporal feature. Experimental results on NTU RGB+D 60, NTU RGB+D 120, NW-UCLA and PKU-MMD I datasets demonstrate significant improvements in distinguishing ambiguous actions.

Figures

Figures reproduced from arXiv: 2412.14833 by the authors.

Figure 1
Figure 1. The overall framework of our proposed method. [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Adaptive Cross-Dimensional Feature Aggregation(AC-FA) module Adaptive Channel Dimension Aggregation(ACDA) [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Accuracy of 23 ambiguous actions (9 groups) in descending order. Group 1: writing, typing on a keyboard, playing with phone, reading; Group 2: jump up, hopping; Group 3: clapping, rub two hands together, tear up paper; Group 4: take off jacket, wear jacket; Group 5: shake head, nod head / bow; Group 6: standing up, sitting down; Group 7: salute, taking a selfie, brushing hair, hand waving; Group 8: take off a shoe, … view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Adaptive learning matrix of the action "salute" on NTU-RGB+D 60 dataset [PITH_FULL_IMAGE:figures/full_fig_p018_4.png]
Figure 5
Figure 5. Figure 5: Representations of long-term action sequences in the NTU-RGB + D 120 Xset dataset using only the backbone (left), backbone + our module w.o. F-RCL (middle), and backbone + our module (right). Each color represents one unique feature. ments. Therefore, the effectiveness…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

61 extracted references · 58 canonical work pages

  1. [1]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Abdelfattah, M., Hassan, M., Alahi, A.: Maskclr: Attention-guided contrastive learning for robust action representation learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 18678– 18687 (2024) 4

    work page 2024

  2. [2]

    Multimedia Systems 30(3), 153 (2024) 2, 4

    Chen, H., He, S., Chen, Z.: Dstc-net: differential spatio-temporal correlation net- work for similar action recognition. Multimedia Systems 30(3), 153 (2024) 2, 4

    work page 2024

  3. [3]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Chen, Y., Zhang, Z., Yuan, C., Li, B., Deng, Y., Hu, W.: Channel-wise topology refinement graph convolution for skeleton-based action recognition. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 13359–13368 (2021) 2, 4, 5, 13, 16, 19, 20

    work page 2021

  4. [4]

    In: Proceedings of the AAAI conference on artificial intelligence

    Chen, Z., Li, S., Yang, B., Li, Q., Liu, H.: Multi-scale spatial temporal graph convolutional network for skeleton-based action recognition. In: Proceedings of the AAAI conference on artificial intelligence. vol. 35, pp. 1113–1122 (2021) 20

    work page 2021

  5. [5]

    In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXIV 16

    Cheng,K.,Zhang,Y.,Cao,C.,Shi,L.,Cheng,J.,Lu,H.:Decouplinggcnwithdrop- graph module for skeleton-based action recognition. In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXIV 16. pp. 536–553. Springer (2020) 20

    work page 2020

  6. [6]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Cheng, K., Zhang, Y., He, X., Chen, W., Cheng, J., Lu, H.: Skeleton-based ac- tion recognition with shift graph convolutional network. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 183–192 (2020) 20

    work page 2020

  7. [7]

    IEEE Transactions on Image Processing 30, 7333–7348 (2021) 20

    Cheng, K., Zhang, Y., He, X., Cheng, J., Lu, H.: Extremely lightweight skeleton- based action recognition with shiftgcn++. IEEE Transactions on Image Processing 30, 7333–7348 (2021) 20

    work page 2021

  8. [8]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Chi, H.g., Ha, M.H., Chi, S., Lee, S.W., Huang, Q., Ramani, K.: Infogcn: Repre- sentation learning for human skeleton-based action recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 20186– 20196 (2022) 4, 6, 16, 19, 20 22 H. Huang et al

    work page 2022

Show all 61 references
  1. [9]

    In: Proceedings of the IEEE conference on com- puter vision and pattern recognition

    Choutas, V., Weinzaepfel, P., Revaud, J., Schmid, C.: Potion: Pose motion repre- sentation for action recognition. In: Proceedings of the IEEE conference on com- puter vision and pattern recognition. pp. 7024–7033 (2018) 2, 3

    2018

  2. [10]

    In: European Conference on Computer Vision

    Do, J., Kim, M.: Skateformer: skeletal-temporal transformer for human action recognition. In: European Conference on Computer Vision. pp. 401–420. Springer (2025) 19, 20

    2025

  3. [11]

    In: Proceedings of the European conference on computer vision (ECCV)

    Dubey, A., Gupta, O., Guo, P., Raskar, R., Farrell, R., Naik, N.: Pairwise confusion for fine-grained visual classification. In: Proceedings of the European conference on computer vision (ECCV). pp. 70–86 (2018) 4

    2018

  4. [12]

    IEEE Transactions on Circuits and Systems for Video Technology28(3), 807–811 (2016) 2

    Hou, Y., Li, Z., Wang, P., Li, W.: Skeleton optical spectra-based action recognition using convolutional neural networks. IEEE Transactions on Circuits and Systems for Video Technology28(3), 807–811 (2016) 2

    2016

  5. [13]

    In: Proceedings of the 2023 5th International Conference on Big-data Service and Intelligent Computa- tion

    Huang, B., Li, X.: Integrating dual-stream cross fusion and ambiguous exclude contrastive learning for enhanced human action recognition. In: Proceedings of the 2023 5th International Conference on Big-data Service and Intelligent Computa- tion. pp. 75–81 (2023) 2, 4

    2023

  6. [14]

    In: Proceedings of the AAAI conference on artificial intelligence

    Ke, L., Peng, K.C., Lyu, S.: Towards to-at spatio-temporal focus for skeleton-based action recognition. In: Proceedings of the AAAI conference on artificial intelligence. vol. 36, pp. 1131–1139 (2022) 20

    2022

  7. [15]

    IEEE signal processing letters24(6), 731– 735 (2017) 2

    Ke, Q., An, S., Bennamoun, M., Sohel, F., Boussaid, F.: Skeletonnet: Mining deep part features for 3-d action recognition. IEEE signal processing letters24(6), 731– 735 (2017) 2

    2017

  8. [16]

    In: Proceedings of the IEEE confer- ence on computer vision and pattern recognition

    Ke, Q., Bennamoun, M., An, S., Sohel, F., Boussaid, F.: A new representation of skeleton sequences for 3d action recognition. In: Proceedings of the IEEE confer- ence on computer vision and pattern recognition. pp. 3288–3297 (2017) 2

    2017

  9. [17]

    In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XX 16

    Korban, M., Li, X.: Ddgcn: A dynamic directed graph convolutional network for action recognition. In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XX 16. pp. 761–776. Springer (2020) 20

    2020

  10. [18]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Lee, J., Lee, M., Lee, D., Lee, S.: Hierarchically decomposed graph convolutional networks for skeleton-based action recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 10444–10453 (2023) 2, 4, 6, 16, 19, 20

    2023

  11. [19]

    Li, C., Hou, Y., Wang, P., Li, W.: Joint distance maps based action recognition withconvolutional neuralnetworks.IEEESignal ProcessingLetters 24(5),624–628 (2017) 2

    2017

  12. [20]

    Li, M., Chen, S., Chen, X., Zhang, Y., Wang, Y., Tian, Q.: Actional-structural graphconvolutionalnetworksforskeleton-basedactionrecognition.In:Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 3595–3603 (2019) 3, 20

    2019

  13. [21]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Li, S., Li, W., Cook, C., Zhu, C., Gao, Y.: Independently recurrent neural network (indrnn): Building a longer and deeper rnn. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 5457–5466 (2018) 2

    2018

  14. [22]

    In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, Au- gust 23–28, 2020, Proceedings, Part VI 16

    Li, X., Shuai, B., Tighe, J.: Directional temporal modeling for action recognition. In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, Au- gust 23–28, 2020, Proceedings, Part VI 16. pp. 275–291. Springer (2020) 3

    2020

  15. [23]

    In: International Conference on Neural Information Processing

    Liang, X., Wang, X., Lei, Z., Liao, S., Li, S.Z.: Soft-margin softmax for deep classi- fication. In: International Conference on Neural Information Processing. pp. 413–

  16. [24]

    In: Proceedings of the 28th ACM international conference on multimedia

    Lin, L., Song, S., Yang, W., Liu, J.: Ms2l: Multi-task self-supervised learning for skeleton based action recognition. In: Proceedings of the 28th ACM international conference on multimedia. pp. 2490–2498 (2020) 20

    2020

  17. [25]

    Liu,C.,Hu,Y.,Li,Y.,Song,S.,Liu,J.:Pku-mmd:Alargescalebenchmarkforcon- tinuousmulti-modalhumanactionunderstanding.arXivpreprintarXiv:1703.07475 (2017) 12

    2017 arXiv

  18. [26]

    arXiv preprint arXiv:2411.18941 (2024) 16, 20

    Liu, H., Liu, Y., Ren, M., Wang, H., Wang, Y., Sun, Z.: Revealing key details to see differences: A novel prototypical perspective for skeleton-based action recognition. arXiv preprint arXiv:2411.18941 (2024) 16, 20

    2024 arXiv

  19. [27]

    IEEE trans- actions on pattern analysis and machine intelligence42(10), 2684–2701 (2019) 12

    Liu, J., Shahroudy, A., Perez, M., Wang, G., Duan, L.Y., Kot, A.C.: Ntu rgb+ d 120: A large-scale benchmark for 3d human activity understanding. IEEE trans- actions on pattern analysis and machine intelligence42(10), 2684–2701 (2019) 12

    2019

  20. [28]

    IEEE transactions on pattern analysis and machine intelligence40(12), 3007–3021 (2017) 2

    Liu, J., Shahroudy, A., Xu, D., Kot, A.C., Wang, G.: Skeleton-based action recog- nition using spatio-temporal lstm network with trust gates. IEEE transactions on pattern analysis and machine intelligence40(12), 3007–3021 (2017) 2

    2017

  21. [29]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Liu, Z., Zhang, H., Chen, Z., Wang, Z., Ouyang, W.: Disentangling and unifying graph convolutions for skeleton-based action recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 143–152 (2020) 20

    2020

  22. [30]

    Journal of machine learning research 9(11) (2008) 18

    Van der Maaten, L., Hinton, G.: Visualizing data using t-sne. Journal of machine learning research 9(11) (2008) 18

    2008

  23. [31]

    In: Proceedings of the IEEE/CVF winter confer- ence on applications of computer vision

    Misra, D., Nalamada, T., Arasanipalai, A.U., Hou, Q.: Rotate to attend: Convolu- tional triplet attention module. In: Proceedings of the IEEE/CVF winter confer- ence on applications of computer vision. pp. 3139–3148 (2021) 7

    2021

  24. [32]

    IEEE Transactions on Image Pro- cessing 33, 2477–2490 (2024) 4

    Myung, W., Su, N., Xue, J.H., Wang, G.: Degcn: Deformable graph convolutional networks for skeleton-based action recognition. IEEE Transactions on Image Pro- cessing 33, 2477–2490 (2024) 4

    2024

  25. [33]

    Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z., Lin, Z., Desmaison, A., Antiga, L., Lerer, A.: Automatic differentiation in pytorch (2017) 13

    2017

  26. [34]

    Computer Vision and Image Understanding150, 109–125 (2016) 3

    Peng, X., Wang, L., Wang, X., Qiao, Y.: Bag of visual words and fusion methods for action recognition: Comprehensive study and good practice. Computer Vision and Image Understanding150, 109–125 (2016) 3

    2016

  27. [35]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Qu, H., Cai, Y., Liu, J.: Llms are good action recognizers. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 18395– 18406 (2024) 4

    2024

  28. [36]

    arXiv preprint arXiv:1609.04747 (2016) 13

    Ruder, S.: An overview of gradient descent optimization algorithms. arXiv preprint arXiv:1609.04747 (2016) 13

    2016 arXiv

  29. [37]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Shahroudy, A., Liu, J., Ng, T.T., Wang, G.: Ntu rgb+ d: A large scale dataset for 3d human activity analysis. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 1010–1019 (2016) 12

    2016

  30. [38]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Shi, L., Zhang, Y., Cheng, J., Lu, H.: Two-stream adaptive graph convolutional networks for skeleton-based action recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 12026–12035 (2019) 2, 3, 16, 20

    2019

  31. [39]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Si, C., Chen, W., Wang, W., Wang, L., Tan, T.: An attention enhanced graph convolutional lstm network for skeleton-based action recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 1227–1236 (2019) 20 24 H. Huang et al

    2019

  32. [40]

    In: Proceedings of the AAAI conference on artificial intelligence

    Song, S., Lan, C., Xing, J., Zeng, W., Liu, J.: An end-to-end spatio-temporal attention model for human action recognition from skeleton data. In: Proceedings of the AAAI conference on artificial intelligence. vol. 31 (2017) 2

    2017

  33. [41]

    IEEE transactions on pattern analysis and machine intelligence45(2), 1474–1488 (2022) 20

    Song, Y.F., Zhang, Z., Shan, C., Wang, L.: Constructing stronger and faster base- lines for skeleton-based action recognition. IEEE transactions on pattern analysis and machine intelligence45(2), 1474–1488 (2022) 20

    2022

  34. [42]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Su, K., Liu, X., Shlizerman, E.: Predict & cluster: Unsupervised skeleton based ac- tion recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 9631–9640 (2020) 3

    2020

  35. [43]

    In: Proceedings of the IEEE International Conference on Computer Vision (ICCV) (December 2015) 4

    Veeriah, V., Zhuang, N., Qi, G.J.: Differential recurrent neural networks for action recognition. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV) (December 2015) 4

    2015

  36. [44]

    In: Pro- ceedings of the AAAI Conference on Artificial Intelligence

    Wang, H., Chu, H., Siming, F., Liu, Z., Hu, H.: Renovate yourself: Calibrating feature representation of misclassified pixels for semantic segmentation. In: Pro- ceedings of the AAAI Conference on Artificial Intelligence. vol. 36, pp. 2450–2458 (2022) 10

    2022

  37. [45]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Wang, J., Nie, X., Xia, Y., Wu, Y., Zhu, S.C.: Cross-view action modeling, learning and recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 2649–2656 (2014) 12

    2014

  38. [46]

    In: Proceedings of the European conference on computer vision (ECCV)

    Wu, Y., He, K.: Group normalization. In: Proceedings of the European conference on computer vision (ECCV). pp. 3–19 (2018) 7

    2018

  39. [47]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Xiang, W., Li, C., Zhou, Y., Wang, B., Zhang, L.: Generative action description prompts for skeleton-based action recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 10276–10285 (2023) 20

    2023

  40. [48]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Xie, J., Meng, Y., Zhao, Y., Nguyen, A., Yang, X., Zheng, Y.: Dynamic semantic- based spatial graph convolution network for skeleton-based human action recog- nition. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 38, pp. 6225–6233 (2024) 20

    2024

  41. [49]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Xu,K.,Ye,F.,Zhong,Q.,Xie,D.:Topology-awareconvolutionalneuralnetworkfor efficient skeleton-based action recognition. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 36, pp. 2866–2874 (2022) 20

    2022

  42. [50]

    In: Proceedings of the ieee/cvf conference on computer vision and pattern recognition

    Yan, A., Wang, Y., Li, Z., Qiao, Y.: Pa3d: Pose-action 3d machine for video recog- nition. In: Proceedings of the ieee/cvf conference on computer vision and pattern recognition. pp. 7922–7931 (2019) 2, 3

    2019

  43. [51]

    In: Proceedings of the AAAI conference on ar- tificial intelligence

    Yan, S., Xiong, Y., Lin, D.: Spatial temporal graph convolutional networks for skeleton-based action recognition. In: Proceedings of the AAAI conference on ar- tificial intelligence. vol. 32 (2018) 2, 3, 16, 20

    2018

  44. [52]

    In: Proceedings of the 28th ACM international conference on multimedia

    Ye, F., Pu, S., Zhong, Q., Li, C., Xie, D., Tang, H.: Dynamic gcn: Context-enriched topology learning for skeleton-based action recognition. In: Proceedings of the 28th ACM international conference on multimedia. pp. 55–63 (2020) 20

    2020

  45. [53]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Zeng, A., Sun, X., Yang, L., Zhao, N., Liu, M., Xu, Q.: Learning skeletal graph neural networks for hard 3d pose estimation. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 11436–11445 (2021) 20

    2021

  46. [54]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Zhang, H., Liu, D., Xiong, Z.: Two-stream action recognition-oriented video super- resolution. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 8799–8808 (2019) 3

    2019

  47. [55]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Zhang, J., Lin, L., Liu, J.: Hierarchical consistent contrastive learning for skeleton- based action recognition with growing augmentations. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 37, pp. 3427–3435 (2023) 20 Title Suppressed Due to Excessive Length 25

    2023

  48. [56]

    In: proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Zhang, P., Lan, C., Zeng, W., Xing, J., Xue, J., Zheng, N.: Semantics-guided neural networks for efficient skeleton-based human action recognition. In: proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 1112– 1121 (2020) 20

    2020

  49. [57]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Zhou, H., Liu, Q., Wang, Y.: Learning discriminative representations for skele- ton based action recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 10608–10617 (2023) 2, 4

    2023

  50. [58]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Zhou, Y., Yan, X., Cheng, Z.Q., Yan, Y., Dai, Q., Hua, X.S.: Blockgcn: Redefine topology awareness for skeleton-based action recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 2049– 2058 (2024) 4, 6, 16, 20

    2024

  51. [59]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Zhu, Y., Han, H., Yu, Z., Liu, G.: Modeling the relative visual tempo for self- supervised skeleton-based action recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 13913–13922 (2023) 19, 20

    2023

  52. [60]

    Zhuang, P., Wang, Y., Qiao, Y.: Learning attentive pairwise interaction for fine- grainedclassification.ProceedingsoftheAAAIConferenceonArtificialIntelligence 34(07), 13130–13137 (2020) 4

    2020

  53. [421]

    Springer (2017) 10 Title Suppressed Due to Excessive Length 23

    2017

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.