Research map / 3D vision / Face and shape
SHERF: Generalizable Human NeRF from a Single Image
Shoukang Hu, Fangzhou Hong, Liang Pan, Haiyi Mei, Lei Yang, Ziwei Liu
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 point-level features to global features (ablation) | SHERF · THuman novel view | PSNR: 22.38 -> 23.26 | Improved |
| add pixel-aligned features to global features (ablation) | SHERF · THuman novel view | PSNR: 22.38 -> 23.72 | Improved |
| add global feature to point-level and pixel-aligned (ablation) | SHERF · THuman novel view | PSNR: 24.08 -> 24.44 | Improved |
| add feature fusion transformer (ablation) | SHERF · THuman novel view | PSNR: 24.44 -> 24.66 | Improved |
| add feature fusion transformer (ablation) | SHERF · THuman novel pose | PSNR: 24.08 -> 24.26 | Improved |
| use transformer fusion instead of AdaIN (ablation) | SHERF · RenderPeople novel view | PSNR: 22.38 -> 22.88 | Improved |
| use transformer fusion instead of SIREN (ablation) | SHERF · RenderPeople novel view | PSNR: 21.62 -> 22.88 | Improved |
| train with free view inputs instead of front view only (ablation) | SHERF · THuman novel view, free view test | PSNR: 22.60 -> 24.66 | Improved |
| use estimated SMPL and camera instead of ground truth | SHERF · RenderPeople novel view | PSNR: 22.88 -> 19.64 | Hurt |
| evaluate cross-dataset (train THuman, test ZJU_MoCap) | SHERF vs NHP · ZJU_MoCap novel view | PSNR: NHP 22.07 vs SHERF 21.87 | Mixed |
Later papers that compared with this method
3 later accepted papers report SHERF as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Feed-forward Human Performance Capture via Progressive Canonical Space Updates (ICLR 2026 · better on 3 of 3 shared results)
- Generalizable Human Gaussians from Single-View Image (ICLR 2025 · better on 3 of 3 shared results)
- Disco4D: Disentangled 4D Human Generation and Animation from a Single Image (CVPR 2025 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MPS-NeRF — MPS-NeRF: Generalizable 3D Human Rendering From Multiview Images (TPAMI 2022) · better on 9 of 9 shared results
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
- PoseGen: Learning to Generate 3D Human Pose Dataset with NeRF (AAAI 2024 · similarity 0.88)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.87)
- TFS-NeRF: Template-Free NeRF for Semantic 3D Reconstruction of Dynamic Scene (NeurIPS 2024 · similarity 0.87)
- GeneMAN: Generalizable Single-Image 3D Human Reconstruction from Multi-Source Human Data (NeurIPS 2025 · similarity 0.87)
- NeRFDeformer: NeRF Transformation from a Single View via 3D Scene Flows (CVPR 2024 · similarity 0.87)
- Guess The Unseen: Dynamic 3D Scene Reconstruction from Partial 2D Glimpses (CVPR 2024 · 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.
