</> Code with Papers by Syntology

Papers › Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimation

Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimation

16 Aug 2021ICCV 2021 10arXiv:2108.07181archive 2025-07-28

Ailing Zeng, Xiao Sun, Lei Yang, Nanxuan Zhao, Minhao Liu, Qiang Xu

Various deep learning techniques have been proposed to solve the single-view 2D-to-3D pose estimation problem. While the average prediction accuracy has been improved significantly over the years, the performance on hard poses with depth ambiguity, self-occlusion, and complex or rare poses is still far from satisfactory. In this work, we target these hard poses and present a novel skeletal GNN learning solution. To be specific, we propose a hop-aware hierarchical channel-squeezing fusion layer to effectively extract relevant information from neighboring nodes while suppressing undesired noises in GNN learning. In addition, we propose a temporal-aware dynamic graph construction procedure that is robust and effective for 3D pose estimation. Experimental results on the Human3.6M dataset show that our solution achieves 10.3\% average prediction accuracy improvement and greatly improves on hard poses over state-of-the-art techniques. We further apply the proposed technique on the skeleton-based action recognition task and also achieve state-of-the-art performance. Our code is available at https://github.com/ailingzengzzz/Skeletal-GNN.

PaperPDFConference PDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

3D Human Pose Estimation3D Pose EstimationAction RecognitionPose EstimationSkeleton Based Action Recognitiongraph construction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Human Pose Estimation Human3.6M Skeletal GNN Average MPJPE (mm) 47.9 #52 of 88 Archive leaderboard report
3D Human Pose Estimation Human3.6M Skeletal GNN Multi-View or Monocular Monocular #52 of 88 Archive leaderboard report
3D Human Pose Estimation Human3.6M Skeletal GNN Using 2D ground-truth joints No #52 of 88 Archive leaderboard report
3D Human Pose Estimation MPI-INF-3DHP Skeletal GNN AUC 46.2 #91 of 108 Archive leaderboard report
3D Human Pose Estimation MPI-INF-3DHP Skeletal GNN PCK 82.1 #91 of 108 Archive leaderboard report
Skeleton Based Action Recognition NTU RGB+D Skeletal GNN Accuracy (CS) 91.6 #37 of 135 Archive leaderboard report
Skeleton Based Action Recognition NTU RGB+D Skeletal GNN Accuracy (CV) 96.7 #37 of 135 Archive leaderboard report
Skeleton Based Action Recognition NTU RGB+D Skeletal GNN Ensembled Modalities 4 #37 of 135 Archive leaderboard report
Skeleton Based Action Recognition NTU RGB+D 120 Skeletal GNN Accuracy (Cross-Setup) 89.2 #32 of 83 Archive leaderboard report
Skeleton Based Action Recognition NTU RGB+D 120 Skeletal GNN Accuracy (Cross-Subject) 87.5 #32 of 83 Archive leaderboard report
Skeleton Based Action Recognition NTU RGB+D 120 Skeletal GNN Ensembled Modalities 4 #32 of 83 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections