Decanus to Legatus: Synthetic training for 2D-3D human pose lifting
Not on the map. Why?
Abstract (arXiv)
3D human pose estimation is a challenging task because of the difficulty to acquire ground-truth data outside of controlled environments. A number of further issues have been hindering progress in building a universal and robust model for this task, including domain gaps between different datasets, unseen actions between train and test datasets, various hardware settings and high cost of annotation, etc. In this paper, we propose an algorithm to generate infinite 3D synthetic human poses (Legatus) from a 3D pose distribution based on 10 initial handcrafted 3D poses (Decanus) during the training of a 2D to 3D human pose lifter neural network. Our results show that we can achieve 3D pose estimation performance comparable to methods using real data from specialized datasets but in a zero-shot setup, showing the generalization potential of our framework.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 29 of its 52 references that are in the map. First posted on arXiv 5 Oct 2022.
Builds on
29 of 52 references.
References · 29
- 2014Microsoft COCO: Common Objects in Contextcited by 17,608
- 2019Expressive Body Capture: 3D Hands, Face, and Body from a Single Imagecited by 1,612
- 2016Keep it SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Imagecited by 695
- 2021Action-Conditioned 3D Human Motion Synthesis with Transformer VAEcited by 501
- 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 494
- 2020Reliable Fidelity and Diversity Metrics for Generative Modelscited by 434
- 2017A simple yet effective baseline for 3d human pose estimationcited by 401
- 2017Learning from Synthetic Humanscited by 359
- 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 221
- 2020Beyond Static Features for Temporally Consistent 3D Human Pose and Shape from a Videocited by 163
- 2020Combining Implicit Function Learning and Parametric Models for 3D Human Reconstructioncited by 133
- 2018Pose2Seg: Detection Free Human Instance Segmentationcited by 107
17 more
- 2021Graph Stacked Hourglass Networks for 3D Human Pose Estimationcited by 95
- 2020Weakly Supervised 3D Human Pose and Shape Reconstruction with Normalizing Flowscited by 94
- 2021PoseAug: A Differentiable Pose Augmentation Framework for 3D Human Pose Estimationcited by 89
- 2020Cascaded deep monocular 3D human pose estimation with evolutionary training datacited by 80
- 2018Learning Monocular 3D Human Pose Estimation from Multi-view Imagescited by 70
- 2020Weakly-Supervised 3D Human Pose Learning via Multi-view Images in the Wildcited by 66
- 2020CanonPose: Self-Supervised Monocular 3D Human Pose Estimation in the Wildcited by 65
- 20193DPeople: Modeling the Geometry of Dressed Humanscited by 58
- 2015Sparseness Meets Deepness: 3D Human Pose Estimation from Monocular Videocited by 54
- 2021Reconstructing 3D Human Pose by Watching Humans in the Mirrorcited by 47
- 2021Context Modeling in 3D Human Pose Estimation: A Unified Perspectivecited by 37
- 2020Predicting Camera Viewpoint Improves Cross-dataset Generalization for 3D Human Pose Estimationcited by 36
- 2020Bodies at Rest: 3D Human Pose and Shape Estimation from a Pressure Image using Synthetic Datacited by 34
- 2019Multiview-Consistent Semi-Supervised Learning for 3D Human Pose Estimationcited by 31
- 2021Unsupervised Human Pose Estimation through Transforming Shape Templatescited by 26
- 2020Deep Reinforcement Learning for Active Human Pose Estimationcited by 7
- 2016Learning camera viewpoint using CNN to improve 3D body pose estimationcited by 6
Built on by
Citing papers per year: 2023: 1.
1 paper.