Improving Skeleton-based Action Recognition with Interactive Object Information

Fengli Shen; Hao Wen; Jialin Cui; Zhe-Ming Lu; Ziqian Lu

REVIEW 3 major objections 5 minor 47 references

Improving Skeleton-based Action Recognition with Interactive Object Information

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Detected interactive objects inserted into the skeleton graph push NTU RGB+D 60 accuracy to 96.7% cross-subject and 99.2% cross-view, surpassing prior state of the art.

desk verdict Plausible idea with a real SOTA claim on NTU60, but the unspecified self-training split and an internal data inconsistency keep me from trusting the numbers as-is. read the letter →

arxiv 2501.05066 v1 pith:RJUREWIC submitted 2025-01-09 cs.CV cs.AI

Hao Wen , Ziqian Lu , Fengli Shen , Zhe-Ming Lu , Jialin Cui This is my paper · ORCID

classification cs.CVcs.AI

keywords graphconvolutionalnetworksskeleton-basedactionrecognitionhuman-objectinteractionvariableobjectdetectionpseudo-labelsdataaugmentationNTURGB+DJXGC24

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 the reason skeleton-only action recognition confuses actions such as reading and writing is that the skeleton graph carries no information about the objects a person touches. To fix this, it adds detected interactive objects as extra graph nodes, each carrying position, detection confidence, and a text-derived class embedding, and builds a Spatial Temporal Variable Graph Convolutional Network (ST-VGCN) that can process a variable number of nodes per frame. The authors construct two object-enriched datasets (NTU RGB+D+Object 60 and JXGC 24) with more than two million object-node annotations and report state-of-the-art accuracy, 96.7% on NTU RGB+D 60 cross-subject and 99.2% on cross-view. A sympathetic reader would care because the gains come from a cheap modality, object detection, applied to a privacy-friendly skeleton representation, and because the variable-graph machinery removes a fixed-topology constraint of prior GCNs.

What carries the argument

The load-bearing object is the Variable Graph (VG), built under the rule 'one frame, one graph': every frame contributes all detected skeletons and all detected objects as a single directed node set, so the graph size varies across frames and videos. A second mechanism is the object node itself, whose three concatenated parts, position, detector probability, and a CLIP text-encoded class attribute, carry the interactive information into the network. ST-VGCN then handles the variable size with node padding and keeps object information from dominating through Class Attribute Fusion, Weighted Node Pooling, Node Balance Loss, and Random Node Attack augmentation. The unidirectional object-to-skeleton edge is what prevents padded empty nodes from turning into noise during graph convolution.

What would settle it

Train the same ST-VGCN with object nodes generated by a detector trained only on the official training split of NTU RGB+D 60, with all validation and test videos excluded from the unlabeled pool, and rerun the X-Sub and X-View benchmarks; if the reported 96.7% and 99.2% cannot be reproduced without test-frame leakage, the claimed gains are not from object information alone. A simpler check is to count how often the detected object category on test frames matches the ground-truth action category for actions like reading, and whether removing those object categories from the annotation changes accuracy.

Watch

Extended reading notes

Core claim

On its own terms, the paper's discovery is that interactive objects are not auxiliary context but a missing part of the action's state, and that injecting them into the skeleton graph as additional nodes is enough to surpass prior skeleton-based state of the art. The object node feature is the concatenation of object position, detector probability, and a CLIP-encoded class attribute; each frame becomes a directed Variable Graph with all skeleton joints of all people plus all object nodes, with skeleton-skeleton edges in both directions and object-to-skeleton edges in one direction. ST-VGCN is an ST-GCN-style network augmented with node padding for variable graph sizes, a Class Attribute Fusion module, weighted node pooling, a Node Balance Loss that keeps skeleton and object channels in proportion, and a Random Node Attack augmentation that inserts fake object nodes during training to prevent the classifier from latching onto object identity. With these pieces, the method reports 96.7% (X-Sub) and 99.2% (X-View) on NTU RGB+D 60, beating the strongest cited prior result by 2.6 and 1.4 percentage points, and 91.4% on the new JXGC 24 'Both' split, 8.1 points above its own no-object baseline.

Load-bearing premise

The load-bearing premise is that the roughly two million object-node annotations produced by the self-trained detector are accurate and that the detector never saw test-set frames while being trained or pseudo-labelled; the paper does not specify which videos formed the unlabeled data set, so this contamination can neither be confirmed nor ruled out.

Editorial extensions

If this is right

Actions that look nearly identical in joint coordinates, such as reading versus writing, become separable once a book or pen node is present.
The same fixed skeleton backbones can be upgraded by appending object nodes, at small cost: parameter count rises only from 2.71M to 2.72M and inference time from 6.43 ms to 6.73 ms in the reported setting.
Recognition in industrial settings can be organized around two independent dimensions, what action and which workpiece, and object nodes improve both, with the largest jump on workpiece classification (99.8% vs. 93.0% for the strongest prior baseline in the 'Workpiece' split).
Random Node Attack makes the model robust to spurious objects: without it, adding a book node to an unrelated video collapses accuracy to 58.7%; with it, accuracy remains at 95.8%.

Reading between the lines

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

Beyond the paper, if the pseudo-label pipeline is sound, the same object-node recipe could be applied to other skeleton benchmarks without manual annotation, since the object detector only needs RGB frames and the method already works with 2D pose.
Beyond the paper, the variable-graph construction is not limited to objects; the paper's own conclusion suggests extending it to human-background and object-object relations, which would let one GCN ingest heterogeneous scene information under the same padding scheme.
Beyond the paper, a natural test is whether Random Node Attack can serve as a general regularizer for any multimodal graph where one modality's identity features are much stronger than another's, not just object-versus-skeleton imbalance.
Beyond the paper, because the NTU RGB+D 120 evaluation reuses object annotations from the first 60 classes, one would expect the advantage to shrink on the 60 new classes; the paper's reported 120-level gains of 1.5 to 1.7 points are an upper bound for generalization to unseen objects.

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.

Desk Editor's Note

read the letter

Short version: this is a well-written paper with a plausible core idea — add object nodes to skeleton graphs and model the variable graph — and the ablations are internally consistent. But the headline NTU60 numbers rest on a self-training pipeline whose train/test separation is never described, and there's a concrete sample-count inconsistency that makes me wary. I'd send it to review, but I'd insist on a stated split and a rerun or release before believing the 96.7/99.2.

What's actually new: the variable graph construction (handles a changing number of object nodes per frame) and Random Node Attack, a data augmentation that injects random object nodes to prevent the model from latching onto category alone. That augmentation is a good idea, and Table 7 shows it works: without it, inserting a 'Book' node nukes accuracy to 58.7; with it, 95.8. That's compelling evidence of the failure mode and the fix. The CLIP-based class attributes and Node Balance Loss are minor but sensible. JXGC 24, though small, is a useful industrial benchmark with a clean three-way split (action, workpiece, both).

The soft spots are real. The big one is Section 3.1.2 and Algorithm 1: the self-training object detector is described, but Dl and Dp are never defined. If Dp includes test frames, the object nodes at test time come from a detector that has seen the test videos, which would inflate the reported gains. I'm not alleging misconduct; I'm saying the paper as written doesn't rule it out. That's load-bearing because the entire contribution is the object-node signal. Also, Table 1 says JXGC 24 has 3,174 samples while Section 4.1.3 says 1,770; those can't both be right. Minor, but it undermines confidence in the numbers. On NTU120, the method doesn't beat prior SOTA, so the abstract's 'surpasses the previous state-of-the-art on multiple benchmarks' is oversold — it's SOTA on two NTU60 splits only. A GitHub link appears, but no statement that code or data are released, so independent verification is currently impossible.

Who is this for? People working on skeleton-based action recognition, especially human-object interaction. The idea is worth engaging with, and Random Node Attack is a practical trick worth stealing. But I wouldn't take the NTU60 numbers at face value until the self-training split is stated and the data inconsistency fixed.

Recommendation: send to peer review with a major-revision request: state the split, correct the sample count, release the annotations or at least the detection pipeline, and temper the SOTA claim.

Referee Report

3 major / 5 minor

Summary. The paper proposes ST-VGCN, a spatial-temporal variable graph convolutional network for skeleton-based action recognition that adds object nodes encoding position, detection probability, and CLIP text-based class attributes to the skeleton graph. It introduces variable graph construction to handle sequences with variable node counts, and four components: Class Attribute Fusion (CAF), Weighted Node Pooling (WNPool), Node Balance Loss, and Random Node Attack. To obtain object annotations, the authors employ a self-training YOLOv5l detector, producing roughly 2 million object nodes and creating two datasets: NTU RGB+D+Object 60 and JXGC 24. The paper reports state-of-the-art accuracy on NTU RGB+D 60 (96.7% cross-subject, 99.2% cross-view) and large gains on JXGC 24.

Significance. If the object annotations are generated without test-set leakage, the paper would make a substantive contribution: it shows that object category and position information can substantially improve skeleton-based action recognition, it introduces a new annotated dataset and an object-augmented version of NTU60, and it addresses the resulting object-category shortcut with a dedicated augmentation. The ablations are systematic, and the module-by-module analysis (Tables 4 and 6) is useful. However, the validity of the headline numbers depends entirely on the self-training annotation pipeline, whose train/test separation is not described; the reported numbers should not be used to compare against prior work until this is resolved. Reproducibility is also weakened by the dataset-count inconsistency and the lack of object-detection accuracy reporting on the pseudo-labels.

major comments (3)

[Section 3.1.2, Algorithm 1] The composition of the labeled set D_l and the pseudo-labeled set D_p in the self-training object detector is never stated. In particular, the paper does not say whether D_p includes any videos from the NTU RGB+D 60 test splits that are later used to evaluate ST-VGCN. If test frames are included in D_p, the object detector would have been trained or pseudo-labeled on those frames, so the object nodes fed to the action recognizer at test time are not produced by a model blind to the evaluation set. This would directly inflate the reported gains (3.3% on X-Sub, 1.2% on X-View) and would make the comparison against prior skeleton-only methods unfair. The authors must state the exact split used for D_l and D_p, and ideally verify that all test-set videos are excluded from D_p; alternatively, they should rerun the experiments with a detector trained only on training-set frames.
[Section 4.1.3 and Table 1] The JXGC 24 dataset is described as containing 1,770 video samples in Section 4.1.3, but Table 1 lists 3,174 samples. This is a direct factual contradiction in the dataset description. Please correct the number, state which value is actually used in the experiments of Tables 2 and 3, and check whether the dataset composition described elsewhere (8 object types, 3 actions, 24 combined classes) is consistent with the reported count.
[Section 3.1.2 and Experiments] No evaluation of the object detector used for pseudo-labeling is reported. The paper gives no mAP or accuracy for the YOLOv5l self-training process, nor any measure of noise in the approximately 2 million auto-generated object nodes. Since the central improvement is attributed to these object nodes, the reader cannot judge whether the reported gains are robust to detector errors. Please report detection performance on a separated validation set and, if possible, analyze the sensitivity of the action-recognition results to pseudo-label noise.

minor comments (5)

[Section 4.3.1] The text says combining position and class attributes yields a 3.5% improvement, but Table 4 shows 3.9% (96.1 vs. 92.2). Please correct the inconsistency.
[Section 4.3.5] The statement that "the network achieves optimal classification performance 93.0% when the parameter λ is set to 0.1" is ambiguous; it should state the split (presumably JXGC 24 "Both") and be reconciled with the 96.1% final accuracy reported in Table 6.
[Algorithm 1] The loop condition "losscount − losscount−1 > c" is undefined at the first iteration because losscount−1 is not initialized. Please clarify the convergence criterion.
[Abstract and Section 4.3.6] There are typos ("designe", "Aandom"), and the phrase "directional interference" should be defined or replaced with a concrete description of the attack.
[Table 2] The formatting of the last rows is garbled (e.g., "MOurs" and "T able") and several entries are concatenated (e.g., "91.91.8 -"); please ensure the table is readable and that all numbers are correctly aligned.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: object nodes are derived from external object detection and text encoding, not from action labels; the central claim is independently evaluated.

full rationale

The paper's central claim is that adding interactive object nodes to skeleton graphs improves action recognition, evaluated on NTU RGB+D 60/120 and JXGC 24. The object nodes are generated by an object detector (YOLOv5l) and CLIP text encoder, both external to the action labels. No equation in the paper defines object nodes in terms of the action recognition output, and no load-bearing self-citation chain appears: the cited prior work on self-training [33,34], CLIP [19], and skeleton GCNs [5,6] is independent of the authors. The NTU60 accuracy is compared against skeleton-only baselines from the same paper and against prior state-of-the-art methods, giving the object-node contribution an independent anchor. The main weakness is the unspecified composition of Dl and Dp in Algorithm 1 for self-training the object detector; if test frames were included in the pseudo-label pool, the reported gains could be inflated. However, this is a data-integrity and audit concern, not circularity: the object-node input is not by construction identical to the action label, and the paper does not exhibit any reduction of a predicted quantity to a fitted parameter. The Random Node Attack experiment (Table 7) shows that object-category information can drive predictions, but that demonstrates the informativeness of the object nodes rather than a circular derivation. The JXGC 24 sample count discrepancy (3,174 in Table 1 vs. 1,770 in Sec. 4.1.3) further complicates audit but does not constitute circular reasoning. For these reasons, the derivation chain is self-contained with respect to circularity, and the score is 0.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central method adds no new physical entities; 'object nodes' are graph nodes filled by an external object detector. The main assumptions are about the reliability of auto-generated annotations and the validity of the evaluation protocol.

free parameters (3)

lambda (Node Balance Loss weight) = 0.1
Tuned on JXGC 24 to maximize accuracy; accuracy drops for higher values (Figure 5).
Random Node Attack node count range = 1 to 3
Chosen manually; number of random object nodes added per frame during training (Section 3.3.5).
Number of object classes annotated in NTU60 = 13
Selected object types (cup, toothbrush, comb, chair, book, pen, paper, shoe, eyeglass, hat, phone, keyboard, watch) from NTU60; this choice affects which actions can benefit.

assumptions (5)

domain assumption YOLOv5l object detections are accurate enough that self-training-generated object nodes faithfully represent interactive objects.
The roughly two million object nodes are generated without human verification (Section 3.1.2).
domain assumption CLIP text embeddings of object descriptions provide useful class attributes for action recognition.
Class attributes are encoded with CLIP and concatenated to nodes (Section 3.1.1).
standard math A graph convolution update as used in ST-GCN generalizes to a variable number of nodes after padding.
The spatial GCN block is reused from ST-GCN [5] with a padded adjacency matrix (Section 3.3).
domain assumption The self-training object detector is trained on splits that do not include evaluation videos.
Never stated in Algorithm 1 or Section 3.1.2; if false, test labels leak through object nodes.
domain assumption For NTU120, applying NTU60 object annotations unchanged to the first 60 classes is a valid way to test generalization.
The 60 new NTU120 classes have no object annotations (note in Table 2), yet the model is evaluated on all 120 classes.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Improving Skeleton-based Action Recognition with Interactive Object Information." pith.science (2026). https://pith.science/paper/RJUREWIC

@misc{pith2026250105066,
  author       = {Pith},
  title        = {Pith review of: Improving Skeleton-based Action Recognition with Interactive Object Information},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RJUREWIC}},
  note         = {Machine review of arXiv:2501.05066}
}

read the original abstract

Human skeleton information is important in skeleton-based action recognition, which provides a simple and efficient way to describe human pose. However, existing skeleton-based methods focus more on the skeleton, ignoring the objects interacting with humans, resulting in poor performance in recognizing actions that involve object interactions. We propose a new action recognition framework introducing object nodes to supplement absent interactive object information. We also propose Spatial Temporal Variable Graph Convolutional Networks (ST-VGCN) to effectively model the Variable Graph (VG) containing object nodes. Specifically, in order to validate the role of interactive object information, by leveraging a simple self-training approach, we establish a new dataset, JXGC 24, and an extended dataset, NTU RGB+D+Object 60, including more than 2 million additional object nodes. At the same time, we designe the Variable Graph construction method to accommodate a variable number of nodes for graph structure. Additionally, we are the first to explore the overfitting issue introduced by incorporating additional object information, and we propose a VG-based data augmentation method to address this issue, called Random Node Attack. Finally, regarding the network structure, we introduce two fusion modules, CAF and WNPool, along with a novel Node Balance Loss, to enhance the comprehensive performance by effectively fusing and balancing skeleton and object node information. Our method surpasses the previous state-of-the-art on multiple skeleton-based action recognition benchmarks. The accuracy of our method on NTU RGB+D 60 cross-subject split is 96.7\%, and on cross-view split, it is 99.2\%.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 28 canonical work pages

[1]

Virtual Reality & Intelligent Hardware 1(1), 84–112 (2019) https://doi.org/10.3724/SP.J.2096-5796.2018.0006

LI, Y., HUANG, J., TIAN, F., W ANG, H.-A., DAI, G.-Z.: Gesture interaction in virtual reality. Virtual Reality & Intelligent Hardware 1(1), 84–112 (2019) https://doi.org/10.3724/SP.J.2096-5796.2018.0006

work page arXiv 2019
[2]

In: 2015 12th International Joint Conference on Computer Science and Software Engineering (JCSSE), pp

Khundam, C.: First person movement control with palm normal and hand gesture interaction in virtual reality. In: 2015 12th International Joint Conference on Computer Science and Software Engineering (JCSSE), pp. 325–330 (2015). https: //doi.org/10.1109/JCSSE.2015.7219818

work page arXiv 2015
[3]

In: ICDSMLA 2019: Proceedings of the 1st International Conference on Data Science, Machine Learning and Applications, pp

Rathod, V., Katragadda, R., Ghanekar, S., Raj, S., Kollipara, P., Anitha Rani, I., Vadivel, A.: Smart surveillance and real-time human action recognition using openpose. In: ICDSMLA 2019: Proceedings of the 1st International Conference on Data Science, Machine Learning and Applications, pp. 504–509 (2020). Springer

work page 2020
[4]

Akyon, F.C., Temizel, A.: Deep Architectures for Content Moderation and Movie Content Rating (2022)

work page 2022
[5]

CoRR abs/1801.07455 (2018) 1801.07455

Yan, S., Xiong, Y., Lin, D.: Spatial temporal graph convolutional networks for skeleton-based action recognition. CoRR abs/1801.07455 (2018) 1801.07455

arXiv 2018
[6]

Duan, H., Wang, J., Chen, K., Lin, D.: PYSKL: Towards Good Practices for Skeleton Action Recognition (2022)

work page 2022
[7]

In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp

Chi, H.-G., Ha, M.H., Chi, S., Lee, S.W., Huang, Q., Ramani, K.: Infogcn: Representation learning for human skeleton-based action recognition. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 20154–20164 (2022). https://doi.org/10.1109/CVPR52688.2022.01955 18

arXiv 2022
[8]

In: Pro- ceedings of the IEEE/CVF International Conference on Computer Vision, pp

Chen, Y., Zhang, Z., Yuan, C., Li, B., Deng, Y., Hu, W.: Channel-wise topology refinement graph convolution for skeleton-based action recognition. In: Pro- ceedings of the IEEE/CVF International Conference on Computer Vision, pp. 13359–13368 (2021)

2021

Show all 47 references

[9]

In: Proceedings of the AAAI Conference on Artificial Intelligence (2021)

Yu, B., Liu, Y., Chan, K.C.C.: Multimodal fusion via teacher-student network for indoor action recognition. In: Proceedings of the AAAI Conference on Artificial Intelligence (2021)

2021
[10]

Expert Systems with Applications 244, 123061 (2024) https://doi.org/10.1016/j.eswa

Cheng, Q., Cheng, J., Liu, Z., Ren, Z., Liu, J.: A dense-sparse complementary net- work for human action recognition based on rgb and skeleton modalities. Expert Systems with Applications 244, 123061 (2024) https://doi.org/10.1016/j.eswa. 2023.123061

2024
[11]

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2959–2968 (2021)

Duan, H., Zhao, Y., Chen, K., Shao, D., Lin, D., Dai, B.: Revisiting skeleton- based action recognition. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2959–2968 (2021)

2021
[12]

IEEE Transactions on Pattern Analysis and Machine Intelligence 45(3), 3522–3538 (2023) https://doi.org/10.1109/TPAMI.2022.3177813

Yu, B.X.B., Liu, Y., Zhang, X., Zhong, S.-h., Chan, K.C.C.: Mmnet: A model- based multimodal network for human action recognition in rgb-d videos. IEEE Transactions on Pattern Analysis and Machine Intelligence 45(3), 3522–3538 (2023) https://doi.org/10.1109/TPAMI.2022.3177813

2023
[13]

CoRR abs/2107.02543 (2021) 2107.02543

Mahmud, H., Morshed, M.M., Hasan, M.K.: A deep-learning-based multimodal depth-aware dynamic hand gesture recognition system. CoRR abs/2107.02543 (2021) 2107.02543

2021
[14]

In: 2020 IEEE 23rd International Conference on Information Fusion (FUSION), pp

Gadzicki, K., Khamsehashari, R., Zetzsche, C.: Early vs late fusion in multimodal convolutional neural networks. In: 2020 IEEE 23rd International Conference on Information Fusion (FUSION), pp. 1–6 (2020). https://doi.org/10.23919/ FUSION45008.2020.9190246

2020
[15]

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

Hachiuma, R., Sato, F., Sekii, T.: Unified keypoint-based action recognition framework via structured keypoint pooling. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 22962–22971 (2023)

2023
[16]

IEEE Transactions on Multimedia 25, 4415–4425 (2023) https://doi.org/10.1109/TMM.2022.3175374

Xu, L., Lan, C., Zeng, W., Lu, C.: Skeleton-based mutually assisted inter- acted object localization and human action recognition. IEEE Transactions on Multimedia 25, 4415–4425 (2023) https://doi.org/10.1109/TMM.2022.3175374

2023
[17]

In: 2019 IEEE Winter Conference on Applications of Computer Vision (W ACV), pp

Kim, S., Yun, K., Park, J., Choi, J.Y.: Skeleton-based action recognition of people handling objects. In: 2019 IEEE Winter Conference on Applications of Computer Vision (W ACV), pp. 61–70 (2019). https://doi.org/10.1109/W ACV.2019.00014

2019
[18]

19 2023 International Joint Conference on Neural Networks (IJCNN), 01–09 (2023)

Aganian, D., K¨ ohler, M., Baake, S., Eisenbach, M., Groß, H.-M.: How object information improves skeleton-based human action recognition in assembly tasks. 19 2023 International Joint Conference on Neural Networks (IJCNN), 01–09 (2023)

2023
[19]

In: Meila, M., Zhang, T

Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., Krueger, G., Sutskever, I.: Learning trans- ferable visual models from natural language supervision. In: Meila, M., Zhang, T. (eds.) Proceedings of the 38th I...

2021
[20]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

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)

2016
[21]

In: CVPR (2017)

Cao, Z., Simon, T., Wei, S.-E., Sheikh, Y.: Realtime multi-person 2d pose estimation using part affinity fields. In: CVPR (2017)

2017
[22]

Duan, H., Wang, J., Chen, K., Lin, D.: DG-STGCN: Dynamic Spatial-Temporal Modeling for Skeleton-based Action Recognition (2022)

2022
[23]

IEEE Transactions on Circuits and Systems for Video Tech- nology 32(12), 8646–8659 (2022) https://doi.org/10.1109/TCSVT.2022.3193574

Zhang, J., Jia, Y., Xie, W., Tu, Z.: Zoom transformer for skeleton-based group activity recognition. IEEE Transactions on Circuits and Systems for Video Tech- nology 32(12), 8646–8659 (2022) https://doi.org/10.1109/TCSVT.2022.3193574

2022
[24]

In: Proceedings of the AAAI Conference on Artificial Intelligence, vol

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

2022
[25]

Lee, J., Lee, M., Lee, D., Lee, S.: Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition (2022)

2022
[26]

In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp

Shi, L., Zhang, Y., Cheng, J., Lu, H.: Two-stream adaptive graph convolutional networks for skeleton-based action recognition. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 12018–12027 (2019). https://doi.org/10.1109/CVPR.2019.01230

2019
[27]

IEEE Transactions on Multimedia 25, 1819–1831 (2023) https://doi.org/10.1109/TMM.2022.3168137

Tu, Z., Zhang, J., Li, H., Chen, Y., Yuan, J.: Joint-bone fusion graph convolutional network for semi-supervised skeleton action recognition. IEEE Transactions on Multimedia 25, 1819–1831 (2023) https://doi.org/10.1109/TMM.2022.3168137

2023
[28]

Pattern Recognition 104, 107356 (2020) https://doi.org/10.1016/j.patcog.2020.107356

Li, J., Xie, X., Pan, Q., Cao, Y., Zhao, Z., Shi, G.: Sgm-net: Skeleton-guided mul- timodal network for action recognition. Pattern Recognition 104, 107356 (2020) https://doi.org/10.1016/j.patcog.2020.107356

2020
[29]

In: DAGM German Conference on Pattern Recognition, pp

Duhme, M., Memmesheimer, R., Paulus, D.: Fusion-gcn: Multimodal action recognition using graph convolutional networks. In: DAGM German Conference on Pattern Recognition, pp. 265–281 (2021). Springer 20

2021
[30]

Xu, H., Gao, Y., Hui, Z., Li, J., Gao, X.: Language Knowledge-Assisted Representation Learning for Skeleton-Based Action Recognition (2023)

2023
[31]

In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp

Zhang, P., Lan, C., Zeng, W., Xing, J., Xue, J., Zheng, N.: Semantics-guided neural networks for efficient skeleton-based human action recognition. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1109–1118 (2020). https://doi.org/10.1109/CVP...

2020
[32]

arXiv preprint arXiv:2107.06779 (2021)

Hu, J., Liu, Y., Zhao, J., Jin, Q.: Mmgcn: Multimodal fusion via deep graph convolution network for emotion recognition in conversation. arXiv preprint arXiv:2107.06779 (2021)

2021 arXiv
[33]

National science review 5(1), 44–53 (2018)

Zhou, Z.-H.: A brief introduction to weakly supervised learning. National science review 5(1), 44–53 (2018)

2018
[34]

In: Workshop on Challenges in Representation Learning, ICML, vol

Lee, D.-H., et al.: Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In: Workshop on Challenges in Representation Learning, ICML, vol. 3, p. 896 (2013). Atlanta

2013
[35]

In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition-workshops, pp

Li, W., Zhang, Z., Liu, Z.: Action recognition based on a bag of 3d points. In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition-workshops, pp. 9–14 (2010). IEEE

2010
[36]

In: Workshops at the Twenty-fifth AAAI Conference on Artificial Intelligence (2011)

Sung, J., Ponce, C., Selman, B., Saxena, A.: Human activity detection from rgbd images. In: Workshops at the Twenty-fifth AAAI Conference on Artificial Intelligence (2011)

2011
[37]

In: 2012 IEEE Conference on Computer Vision and Pattern Recognition, pp

Wang, J., Liu, Z., Wu, Y., Yuan, J.: Mining actionlet ensemble for action recog- nition with depth cameras. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1290–1297 (2012). IEEE

2012
[38]

In: 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, pp

Xia, L., Chen, C.-C., Aggarwal, J.K.: View invariant human action recognition using histograms of 3d joints. In: 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, pp. 20–27 (2012). https: //doi.org/10.1109/CVPR W.2012.6239233

2012
[39]

In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

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)

2014
[40]

In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp

Li, T., Liu, J., Zhang, W., Ni, Y., Wang, W., Li, Z.: Uav-human: A large bench- mark for human behavior understanding with unmanned aerial vehicles. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 16261–16270 (2021). https://doi.org/10.1109/...

2021
[41]

IEEE Trans- actions on Pattern Analysis and Machine Intelligence 42(10), 2684–2701 (2020) https://doi.org/10.1109/tpami.2019.2916873

Liu, J., Shahroudy, A., Perez, M., Wang, G., Duan, L.-Y., Kot, A.C.: Ntu rgb+d 21 120: A large-scale benchmark for 3d human activity understanding. IEEE Trans- actions on Pattern Analysis and Machine Intelligence 42(10), 2684–2701 (2020) https://doi.org/10.1109/tpami.2019.2916873

2020
[42]

arXiv preprint arXiv:2201.00443 (2022)

Zhu, G., Zhang, L., Jiang, Y., Dang, Y., Hou, H., Shen, P., Feng, M., Zhao, X., Miao, Q., Shah, S.A.A., et al.: Scene graph generation: A comprehensive survey. arXiv preprint arXiv:2201.00443 (2022)

2022 arXiv
[43]

CoRR abs/1908.10357 (2019) 1908.10357

Cheng, B., Xiao, B., Wang, J., Shi, H., Huang, T.S., Zhang, L.: Bottom-up higher- resolution networks for multi-person pose estimation. CoRR abs/1908.10357 (2019) 1908.10357

2019 arXiv
[44]

IEEE Transactions on Multimedia (2023)

Liu, J., Wang, X., Wang, C., Gao, Y., Liu, M.: Temporal decoupling graph con- volutional network for skeleton-based gesture recognition. IEEE Transactions on Multimedia (2023)

2023
[45]

IEEE Transactions on Image Processing 29, 9532–9545 (2020) https://doi.org/10.1109/tip.2020

Shi, L., Zhang, Y., Cheng, J., Lu, H.: Skeleton-based action recognition with multi-stream adaptive graph convolutional networks. IEEE Transactions on Image Processing 29, 9532–9545 (2020) https://doi.org/10.1109/tip.2020. 3028207

2020 doi
[46]

IEEE Transactions on Image Processing 33, 2477–2490 (2024) https://doi.org/10.1109/TIP.2024.3378886

Myung, W., Su, N., Xue, J.-H., Wang, G.: Degcn: Deformable graph convolutional networks for skeleton-based action recognition. IEEE Transactions on Image Processing 33, 2477–2490 (2024) https://doi.org/10.1109/TIP.2024.3378886

2024
[47]

In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2024) 22

Zhou, Y., Yan, X., Cheng, Z.-Q., Yan, Y., Dai, Q., Hua, X.-S.: Blockgcn: Redefin- ing topology awareness for skeleton-based action recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2024) 22

2024

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

Virtual Reality & Intelligent Hardware 1(1), 84–112 (2019) https://doi.org/10.3724/SP.J.2096-5796.2018.0006

LI, Y., HUANG, J., TIAN, F., W ANG, H.-A., DAI, G.-Z.: Gesture interaction in virtual reality. Virtual Reality & Intelligent Hardware 1(1), 84–112 (2019) https://doi.org/10.3724/SP.J.2096-5796.2018.0006

work page arXiv 2019

[2] [2]

In: 2015 12th International Joint Conference on Computer Science and Software Engineering (JCSSE), pp

Khundam, C.: First person movement control with palm normal and hand gesture interaction in virtual reality. In: 2015 12th International Joint Conference on Computer Science and Software Engineering (JCSSE), pp. 325–330 (2015). https: //doi.org/10.1109/JCSSE.2015.7219818

work page arXiv 2015

[3] [3]

In: ICDSMLA 2019: Proceedings of the 1st International Conference on Data Science, Machine Learning and Applications, pp

Rathod, V., Katragadda, R., Ghanekar, S., Raj, S., Kollipara, P., Anitha Rani, I., Vadivel, A.: Smart surveillance and real-time human action recognition using openpose. In: ICDSMLA 2019: Proceedings of the 1st International Conference on Data Science, Machine Learning and Applications, pp. 504–509 (2020). Springer

work page 2020

[4] [4]

Akyon, F.C., Temizel, A.: Deep Architectures for Content Moderation and Movie Content Rating (2022)

work page 2022

[5] [5]

CoRR abs/1801.07455 (2018) 1801.07455

Yan, S., Xiong, Y., Lin, D.: Spatial temporal graph convolutional networks for skeleton-based action recognition. CoRR abs/1801.07455 (2018) 1801.07455

arXiv 2018

[6] [6]

Duan, H., Wang, J., Chen, K., Lin, D.: PYSKL: Towards Good Practices for Skeleton Action Recognition (2022)

work page 2022

[7] [7]

In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp

Chi, H.-G., Ha, M.H., Chi, S., Lee, S.W., Huang, Q., Ramani, K.: Infogcn: Representation learning for human skeleton-based action recognition. In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 20154–20164 (2022). https://doi.org/10.1109/CVPR52688.2022.01955 18

arXiv 2022

[8] [8]

In: Pro- ceedings of the IEEE/CVF International Conference on Computer Vision, pp

Chen, Y., Zhang, Z., Yuan, C., Li, B., Deng, Y., Hu, W.: Channel-wise topology refinement graph convolution for skeleton-based action recognition. In: Pro- ceedings of the IEEE/CVF International Conference on Computer Vision, pp. 13359–13368 (2021)

2021