Research map / Detection and segmentation / Open-set and few-shot detection
Breaking Bad: A Dataset for Geometric Fracture and Reassembly
Silvia Sellán, Yun-Chun Chen, Ziyi Wu, Animesh Garg, Alec Jacobson
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 |
|---|---|---|---|
| use DGL (graph network) instead of Global or LSTM | shape assembly baselines · everyday subset | part accuracy: 24.6 -> 31.0 | Improved |
| test on objects with more fractured pieces (21-50 vs 2-20) (ablation) | DGL trained on 2-20 pieces · everyday subset | part accuracy: 31.0 -> 7.5 | Hurt |
| train on 2-50 pieces instead of 2-20 (ablation) | DGL · everyday, test 51-100 pieces | chamfer distance: 23.0 -> 18.1 | Improved |
| train on 2-100 pieces instead of 2-20 (ablation) | DGL · everyday, test 51-100 pieces | chamfer distance: 23.0 -> 13.9 | Improved |
| train on 2-100 pieces instead of 2-20 (ablation) | DGL · everyday, test 2-20 pieces | part accuracy: 31.0 -> 29.4 | Hurt |
| fine-tune from everyday-subset pre-trained model instead of training from scratch (ablation) | DGL · artifact subset | part accuracy: 12.8 -> 19.4 | Improved |
| fine-tune from everyday-subset pre-trained model instead of training from scratch (ablation) | LSTM · artifact subset | part accuracy: 6.6 -> 11.7 | Improved |
| fine-tune from everyday-subset pre-trained model instead of training from scratch (ablation) | Global · artifact subset | part accuracy: 12.7 -> 13.3 | Improved |
| pre-train on everyday plus artifact data rather than everyday only, test on unseen subset (ablation) | DGL · other subset | chamfer distance: 38.5 -> 36.6 | Improved |
| test semantic-assembly methods on geometric fracture reassembly vs semantic assembly | Global, LSTM, DGL · Breaking Bad | performance | Hurt |
Closest papers by meaning
- Fantastic Breaks: A Dataset of Paired 3D Scans of Real-World Broken Objects and Their Complete Counterparts (CVPR 2023 · similarity 0.89)
- Scalable Geometric Fracture Assembly via Co-creation Space among Assemblers (AAAI 2024 · similarity 0.86)
- PuzzleFusion++: Auto-agglomerative 3D Fracture Assembly by Denoise and Verify (ICLR 2025 · similarity 0.85)
- Re-assembling the past: The RePAIR dataset and benchmark for real world 2D and 3D puzzle solving (NeurIPS 2024 · similarity 0.84)
- DiffAssemble: A Unified Graph-Diffusion Model for 2D and 3D Reassembly (CVPR 2024 · similarity 0.84)
- Learning Partonomic 3D Reconstruction from Image Collections (CVPR 2025 · similarity 0.83)
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.
