Research map / 3D vision / Face and shape
In-Hand 3D Object Scanning From an RGB Sequence
Shreyas Hampali, Tomas Hodan, Luan Tran, Lingni Ma, Cem Keskin, Vincent Lepetit
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 |
|---|---|---|---|
| Estimate object shape and camera trajectory jointly without supplied poses | In-Hand 3D Object Scanning · HO-3D, average | RMSE Hausdorff distance (mm): 13.41 -> 4.65 | Improved |
| Estimate shape and poses from RGB rather than use the single-image Ye et al. method | HO-3D · HO-3D, average | RMSE Hausdorff distance (mm): 6.76 -> 4.65 | Improved |
| Estimate poses instead of using ground-truth poses | Ours versus UNISURF · HO-3D, average | PSNR: 31.02 -> 31.01 | No clear effect |
| Estimate poses instead of using ground-truth poses | Ours versus UNISURF · HO-3D, average | SSIM: 0.75 -> 0.77 | Improved |
| Estimate poses instead of using ground-truth poses | Ours versus UNISURF · HO-3D, average | LPIPS: 0.34 -> 0.32 | Improved |
| Add synthetic-depth loss (ablation) | In-Hand 3D Object Scanning · HO-3D MDF14, largest segment | ATE AUC: 4.9 -> 5.8 | Improved |
| Add optical-flow loss (ablation) | In-Hand 3D Object Scanning · HO-3D MDF14, largest segment | ATE AUC: 4.8 -> 5.8 | Improved |
| Add shape-regularization loss (ablation) | In-Hand 3D Object Scanning · HO-3D MDF14, largest segment | ATE AUC: 1.1 -> 5.8 | Improved |
| Add synthetic-depth loss (ablation) | In-Hand 3D Object Scanning · HO-3D SM2, largest segment | ATE AUC: 1.0 -> 8.1 | Improved |
| Add optical-flow loss (ablation) | In-Hand 3D Object Scanning · HO-3D SM2, largest segment | ATE AUC: 2.9 -> 8.1 | Improved |
| Add shape-regularization loss (ablation) | In-Hand 3D Object Scanning · HO-3D SM2, largest segment | ATE AUC: 0.5 -> 8.1 | Improved |
| Split the sequence into overlapping segments, initialize from segment pose estimates, then optimize globally | In-Hand 3D Object Scanning · In-hand RGB sequence reconstruction | convergence | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- UNISURF — UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction (ICCV 2021) · better on 2 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
- Hand-held Object Reconstruction from RGB Video with Dynamic Interaction (CVPR 2025 · similarity 0.90)
- Free-Moving Object Reconstruction and Pose Estimation with Virtual Camera (AAAI 2025 · similarity 0.89)
- In-Hand 3D Object Reconstruction from a Monocular RGB Video (AAAI 2024 · similarity 0.88)
- Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images (AAAI 2024 · similarity 0.87)
- UNOPose: Unseen Object Pose Estimation with an Unposed RGB-D Reference Image (CVPR 2025 · similarity 0.87)
- SchemaPose: RGB-Based Category-Level Pose Estimation with Parametric Category Schema (NeurIPS 2026 · 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.
