Skeleton-Graph: Long-Term 3D Motion Prediction From 2D Observations Using Deep Spatio-Temporal Graph CNNs
Not on the map. Why?
Abstract (arXiv)
Several applications such as autonomous driving, augmented reality and virtual reality require a precise prediction of the 3D human pose. Recently, a new problem was introduced in the field to predict the 3D human poses from observed 2D poses. We propose Skeleton-Graph, a deep spatio-temporal graph CNN model that predicts the future 3D skeleton poses in a single pass from the 2D ones. Unlike prior works, Skeleton-Graph focuses on modeling the interaction between the skeleton joints by exploiting their spatial configuration. This is being achieved by formulating the problem as a graph structure while learning a suitable graph adjacency kernel. By the design, Skeleton-Graph predicts the future 3D poses without divergence in the long-term, unlike prior works. We also introduce a new metric that measures the divergence of predictions in the long term. Our results show an FDE improvement of at least 27% and an ADE of 4% on both the GTA-IM and PROX datasets respectively in comparison with prior works. Also, we are 88% and 93% less divergence on the long-term motion prediction in comparison with prior works on both GTA-IM and PROX datasets. Code is available at https://github.com/abduallahmohamed/Skeleton-Graph.git
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 26 of its 44 references that are in the map. First posted on arXiv 21 Sep 2021.
Builds on
26 of 44 references.
References · 26
- 2017Attention Is All You Needcited by 59,932
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
- 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,447
- 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,199
- 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
- 2018Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networkscited by 687
- 2019DeepGCNs: Can GCNs Go as Deep as CNNs?cited by 527
- 2018Simple Baselines for Human Pose Estimation and Trackingcited by 502
- 2017A simple yet effective baseline for 3d human pose estimationcited by 401
- 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 401
- 2020Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Predictioncited by 324
- 2013DeepPose: Human Pose Estimation via Deep Neural Networkscited by 305
14 more
- 2019Resolving 3D Human Pose Ambiguities with 3D Scene Constraintscited by 294
- 2019Learning Trajectory Dependencies for Human Motion Predictioncited by 245
- 2014Efficient Object Localization Using Convolutional Networkscited by 226
- 2020Long-term Human Motion Prediction with Scene Contextcited by 191
- 2020Dynamic Multiscale Graph Neural Networks for 3D Skeleton-Based Human Motion Predictioncited by 132
- 20163D Human Pose Estimation = 2D Pose Estimation + Matchingcited by 112
- 2016MoCap-guided Data Augmentation for 3D Pose Estimation in the Wildcited by 37
- 2020MotioNet: 3D Human Motion Reconstruction from Monocular Video with Skeleton Consistencycited by 31
- 2020Spatio-Temporal Inception Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 18
- 2018Vehicle Pose and Shape Estimation through Multiple Monocular Visioncited by 13
- 2020GraphTCN: Spatio-Temporal Interaction Modeling for Human Trajectory Predictioncited by 12
- 2020Noticing Motion Patterns: Temporal CNN with a Novel Convolution Operator for Human Trajectory Predictioncited by 8
- 2018Human Motion Capture Using a Dronecited by 7
- 2020Social-STAGE: Spatio-Temporal Multi-Modal Future Trajectory Forecastcited by 6
Built on by
Citing papers per year: 2024: 1; 2025: 0; 2026: 1.
2 papers.