Research map / 3D vision / Face and shape
Self-supervised Learning of Motion Capture
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 self-supervised test-time adaptation to the pretrained model | Motion capture model · SURREAL | surface error (mm): 119.4 -> 74.5 | Helped |
| Add self-supervised test-time adaptation to the pretrained model | Motion capture model · SURREAL | per-joint error (mm): 101.6 -> 64.4 | Helped |
| Add self-supervised test-time adaptation to the pretrained model | Motion capture model · SURREAL | recon. error (mm): 351.3 -> 203.9 | Helped |
| Add self-supervised test-time adaptation to the pretrained model | Motion capture model · H3.6M | per-joint error (mm): 125.6 -> 98.4 | Helped |
| Add self-supervised test-time adaptation to the pretrained model | Motion capture model · H3.6M | recon. error (mm): 303.5 -> 145.8 | Helped |
| Add segmentation reprojection loss to keypoint reprojection loss (ablation) | Self-supervised motion capture model · SURREAL | 3D keypoint reconstruction error | Helped |
| Add dense motion reprojection loss to keypoint and segmentation reprojection losses (ablation) | Self-supervised motion capture model · SURREAL | 3D keypoint reconstruction error | Helped |
Closest papers by meaning
- Capturing Humans in Motion: Temporal-Attentive 3D Human Pose and Shape Estimation From Monocular Video (CVPR 2022 · similarity 0.88)
- Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image Synthesis (CVPR 2020 · similarity 0.88)
- VideoMDM: Towards 3D Human Motion Generation From 2D Supervision (NeurIPS 2026 · similarity 0.87)
- AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers (CVPR 2025 · similarity 0.86)
- Motions as Queries: One-Stage Multi-Person Holistic Human Motion Capture (CVPR 2025 · similarity 0.86)
- Self-Supervised Learning of Motion Concepts by Optimizing Counterfactuals (NeurIPS 2025 · similarity 0.86)
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.