Research map / 3D vision / Face and shape
SHARP: Shape-Aware Reconstruction of People in Loose Clothing
Sai Sagar Jinka, Astitva Srivastava, Chandradeep Pokhariya, Avinash Kumar Sharma, P. J. Narayanan
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 the peel map fusion of residual and auxiliary branches (ablation) | SHARP · 3DHumans | Chamfer distance: 9.98 -> 7.718 | Improved |
| add the smoothness loss L_sm (ablation) | SHARP · 3DHumans | Chamfer distance: 8.3652 -> 7.718 | Improved |
| use ResNet backbone instead of U-Net (ablation) | SHARP · 3DHumans | Chamfer distance: 8.417 -> 7.71 | Improved |
| use ResNet backbone instead of Hourglass (ablation) | SHARP · 3DHumans | Chamfer distance: 15.6 -> 7.71 | Improved |
| increase ResNet blocks from 6 to 18 (ablation) | SHARP · 3DHumans | Chamfer distance: 22.81 -> 7.71 | Improved |
| increase ResNet blocks from 9 to 18 (ablation) | SHARP · 3DHumans | Chamfer distance: 8.9 -> 7.71 | No clear effect |
| replace end-to-end fusion with late feature fusion by concatenation (ablation) | SHARP · 3DHumans | Chamfer distance: 7.71 -> 7.57 | No clear effect |
| replace end-to-end fusion with late feature fusion by averaging (ablation) | SHARP · 3DHumans | Chamfer distance: 7.71 -> 8.82 | Hurt |
| add SMPL prior peel maps to PeeledHuman representation | PeeledHuman · 3DHumans | Chamfer distance: 20.88 -> 7.718 | Improved |
| SHARP versus PaMIR on THUman2.0 | SHARP · THUman2.0 | P2S: 0.00529 -> 0.0049 | Mixed |
Later papers that compared with this method
1 later accepted paper reports SHARP as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- UniSHARP: Universal Sharp Monocular View Synthesis (NeurIPS 2026 · better on 4 of 4 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.
- Geo-PIFu — Geo-PIFu: Geometry and Pixel Aligned Implicit Functions for Single-view Human Reconstruction (NeurIPS 2020) · better on 4 of 4 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
- 3DPeople: Modeling the Geometry of Dressed Humans (ICCV 2019 · similarity 0.91)
- VS: Reconstructing Clothed 3D Human from Single Image via Vertex Shift (CVPR 2024 · similarity 0.91)
- Deep Fashion3D: A Dataset and Benchmark for 3D Garment Reconstruction from Single Images (ECCV 2020 · similarity 0.91)
- Combining Implicit Function Learning and Parametric Models for 3D Human Reconstruction (ECCV 2020 · similarity 0.91)
- MultiGO: Towards Multi-level Geometry Learning for Monocular 3D Textured Human Reconstruction (CVPR 2025 · similarity 0.90)
- D^3-Human: Dynamic Disentangled Digital Human from Monocular Video (CVPR 2025 · similarity 0.89)
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.
