Research map / 3D vision / Face and shape
Batch-based Model Registration for Fast 3D Sherd Reconstruction
Jiepeng Wang, Congyi Zhang, Peng Wang, Xin Li, Peter J. Cobb, Christian Theobalt, Wenping Wang
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 |
|---|---|---|---|
| replace Trimmed ICP / Super 4PCS / FGR with Bilateral Boundary ICP for front-back registration (ablation) | sherd reconstruction pipeline · Sherd dataset (123 fragments) | accuracy (mm): 4.71 -> 0.15 (vs Trimmed ICP) | Improved |
| replace Trimmed ICP / Super 4PCS / FGR with Bilateral Boundary ICP for front-back registration (ablation) | sherd reconstruction pipeline · Sherd dataset (123 fragments) | completeness (%): 18.69 -> 96.00 (vs Trimmed ICP) | Improved |
| restrict baseline registration methods to boundary points (ablation) | FGR, Super-4PCS, Trimmed ICP · one batch of 9 pieces | error (mm): baselines 6.15, 3.87, 4.41 vs BBICP 0.13 | No clear effect |
| use batch matching and BBICP to scan fragments in batches | pipeline · fragments per work hour | throughput: 20 -> 85 per hour (best prior vs ours) | Improved |
| deploy the pipeline in the field instead of the lab | pipeline · excavation site, 26 fragments | accuracy (mm): 0.15 (lab) -> about 0.16 (field) | No clear effect |
| match front and back sides by 2D contour shape descriptors instead of position order | batch matching · batch scanning | matching reliability | Improved |
Closest papers by meaning
- 3DPGS: 3D Probabilistic Graph Search for Archaeological Piece Grouping (AAAI 2025 · similarity 0.84)
- Self-Supervised Large Scale Point Cloud Completion for Archaeological Site Restoration (CVPR 2025 · similarity 0.82)
- Archaeoscape: Bringing Aerial Laser Scanning Archaeology to the Deep Learning Era (NeurIPS 2024 · similarity 0.82)
- Re-assembling the past: The RePAIR dataset and benchmark for real world 2D and 3D puzzle solving (NeurIPS 2024 · similarity 0.82)
- Scalable Geometric Fracture Assembly via Co-creation Space among Assemblers (AAAI 2024 · similarity 0.82)
- PHFormer: Multi-Fragment Assembly Using Proxy-Level Hybrid Transformer (AAAI 2024 · similarity 0.81)
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.
