Research map / 3D vision / Human pose estimation
AvatarPoser: Articulated Full-Body Pose Tracking from Sparse Motion Sensing
Jiaxi Jiang, Paul Streli, Huajian Qiu, Andreas Fender, Larissa Laich, Patrick Snape, Christian Holz
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Stabilizer module (ablation) | AvatarPoser · HDM05 | MPJPE full body (cm): 9.29 -> 8.05 | Improved |
| add Stabilizer module (ablation) | AvatarPoser · HDM05 | MPJPE hand (cm): 2.15 -> 1.86 | Improved |
| compute pelvis translation via kinematic chain instead of regressing it (ablation) | AvatarPoser · HDM05 | MPJPE full body (cm): 8.82 -> 8.05 | Improved |
| add FK module (ablation) | AvatarPoser · HDM05 | MPJPE full body (cm): 8.41 -> 8.05 | Improved |
| add FK module (ablation) | AvatarPoser · HDM05 | MPJRE (deg): 6.24 -> 6.39 | Hurt |
| add IK module (ablation) | AvatarPoser · HDM05 | MPJPE hand (cm): 3.17 -> 1.86 | Improved |
| add IK module (ablation) | AvatarPoser · HDM05 | MPJPE full body (cm): 8.07 -> 8.05 | No clear effect |
| add Stabilizer module (ablation) | AvatarPoser · HDM05 | MPJRE (deg): 6.39 -> 6.39 | No clear effect |
| compare AvatarPoser with VAE-HMD | VAE-HMD · AMASS (three inputs) | MPJPE (cm): 6.83 -> 4.18 | Improved |
| compare AvatarPoser with Final IK on MPJVE | AvatarPoser · CMU (cross-dataset) | MPJVE (cm/s): 56.83 vs 35.76 | Mixed |
Later papers that compared with this method
12 later accepted papers report AvatarPoser as a baseline in their results tables; 12 of them report a better number on most of the results they share with it.
- EgoExoMoCap: Distributed Human Motion Capture via Ego- and Exocentric Body Tracking from Head-Mounted Devices (ECCV 2026 · better on 2 of 2 shared results)
- Towards Real-World Wearable Motion Reconstruction (ECCV 2026 · better on 1 of 1 shared result)
- EgoGVAE: Ego-body Mesh Reconstruction via Guided Variational Autoencoder (ECCV 2026 · better on 1 of 1 shared result)
- UniEgoMotion: A Unified Model for Egocentric Motion Reconstruction, Forecasting, and Generation (ICCV 2025 · better on 3 of 3 shared results)
- Full-Body Motion Reconstruction with Sparse Sensing from Graph Perspective (AAAI 2024 · better on 3 of 3 shared results)
- HMD-Poser: On-Device Real-time Human Motion Tracking from Scalable Sparse Observations (CVPR 2024 · better on 1 of 1 shared result)
- Stratified Avatar Generation from Sparse Observations (CVPR 2024 · better on 1 of 1 shared result)
- MANIKIN: Biomechanically Accurate Neural Inverse Kinematics for Human Motion Estimation (ECCV 2024 · better on 1 of 1 shared result)
4 more not listed.
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- HMD-Poser: On-Device Real-time Human Motion Tracking from Scalable Sparse Observations (CVPR 2024 · similarity 0.90)
- FRAME: Floor-aligned Representation for Avatar Motion from Egocentric Video (CVPR 2025 · similarity 0.90)
- Stratified Avatar Generation from Sparse Observations (CVPR 2024 · similarity 0.89)
- Real-Time Simulated Avatar from Head-Mounted Sensors (CVPR 2024 · similarity 0.89)
- RAM-Avatar: Real-time Photo-Realistic Avatar from Monocular Videos with Full-body Control (CVPR 2024 · similarity 0.88)
- EnvPoser: Environment-aware Realistic Human Motion Estimation from Sparse Observations with Uncertainty Modeling (CVPR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
