Research map / 3D vision / Human pose estimation
HumanNeRF: Free-Viewpoint Rendering of Moving People From Monocular Video
Check 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 |
|---|---|---|---|
| Use HumanNeRF's canonical volume and skeletal/non-rigid motion fields instead of Neural Body | HumanNeRF · ZJU-MoCap subjects 377, 386, 387, 392, 393, 394 | LPIPS*: 40.95 -> 24.06; 46.43 -> 28.99; 59.47 -> 35.58; 53.27 -> 32.12; 59.05 -> 36.72; 54.55 -> 32.89 | Helped |
| Use HumanNeRF instead of Neural Body | HumanNeRF · ZJU-MoCap subject 393 | PSNR: 28.61 -> 28.31 | Hurt |
| Add non-rigid motion to skeletal deformation (ablation) | HumanNeRF · ZJU-MoCap average over 6 sequences | PSNR: 29.81 -> 30.24 | Helped |
| Add non-rigid motion to skeletal deformation (ablation) | HumanNeRF · ZJU-MoCap average over 6 sequences | SSIM: 0.9657 -> 0.9679 | Helped |
| Add non-rigid motion to skeletal deformation (ablation) | HumanNeRF · ZJU-MoCap average over 6 sequences | LPIPS*: 34.17 -> 31.73 | Helped |
| Increase the training sequence from 19 to 112 frames (ablation) | HumanNeRF · ZJU-MoCap subject 392 | PSNR: 30.51 -> 31.01 | Helped |
| Increase the training sequence from 19 to 112 frames (ablation) | HumanNeRF · ZJU-MoCap subject 392 | LPIPS*: 45.17 -> 32.75 | Helped |
| Increase the training sequence beyond 112 frames (ablation) | HumanNeRF · ZJU-MoCap subject 392 | PSNR: 31.01 (112 frames); 31.04 (556 frames) | No clear effect |
| Condition non-rigid motion on the position after skeletal warping (decomposition design 3) (ablation) | HumanNeRF motion field · PeopleSnapshot | novel-view synthesis quality | Helped |
| Delay non-rigid motion optimization and enable it coarse-to-fine (ablation) | HumanNeRF · Unseen-view rendering | generalization quality | Helped |
| Add pose correction (ablation) | HumanNeRF · In-the-wild unseen-view rendering | visual alignment and detail | Helped |
| Add non-rigid deformation (ablation) | HumanNeRF · In-the-wild unseen-view rendering | clothing alignment and shape | Helped |
Closest papers by meaning
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.87)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.87)
- Vid2Avatar-Pro: Authentic Avatar from Videos in the Wild via Universal Prior (CVPR 2025 · similarity 0.87)
- HumanNeRF-SE: A Simple yet Effective Approach to Animate HumanNeRF with Diverse Poses (CVPR 2024 · similarity 0.87)
- Motions as Queries: One-Stage Multi-Person Holistic Human Motion Capture (CVPR 2025 · similarity 0.86)
- Free-viewpoint Human Animation with Pose-correlated Reference Selection (CVPR 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.