Research map / 3D vision / 3D generation
AssemblyBench: Physics-Aware Assembly of Complex Industrial Objects
Danrui Li, Jiahao Zhang, Bernhard Egger, Moitreya Chatterjee, Suhas Lohit, Tim K. Marks, Anoop Cherian
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 trajectory predictor instead of naive baseline (ablation) | AssemblyDyno · GT part orders, simulator | SR (%): 3.57 -> 33.57 | Improved |
| add trajectory predictor instead of naive baseline (ablation) | AssemblyDyno · standard setting, simulator | SR (%): 1.43 -> 13.57 | Improved |
| add text instructions (ablation) | AssemblyDyno · GT part orders, simulator | SR (%): 31.79 -> 33.57 | Improved |
| add text instructions (ablation) | AssemblyDyno · standard setting, simulator | SR (%): 15.00 -> 13.57 | Mixed |
| add text instructions (ablation) | AssemblyDyno · GT part orders, final pose | SR (%): 42.86 -> 44.29 | Improved |
| add point-cloud loss L_P (ablation) | AssemblyDyno · GT part orders, simulator | SR (%): 27.50 -> 33.57 | Improved |
| add translation loss L_T (ablation) | AssemblyDyno · GT part orders, final pose | PA (%): 68.43 -> 79.69 | Improved |
| add rotation loss L_R (ablation) | AssemblyDyno · GT part orders, simulator | SR (%): 26.43 -> 33.57 | Improved |
| add rotation smoothness loss L_SR (ablation) | AssemblyDyno · GT part orders, simulator | SR (%): 5.36 -> 33.57 | Improved |
| raise rotation loss weight to 10 (ablation) | AssemblyDyno · GT part orders, simulator | SR (%): 33.57 -> 32.14 | No clear effect |
| lower rotation smoothness weight to 1 (ablation) | AssemblyDyno · GT part orders, simulator | SR (%): 33.57 -> 20.00 | Hurt |
| mask up to two step diagrams (text kept) | AssemblyDyno · AssemblyBench test | final pose SR (%): 40.36 -> 22.86 | Hurt |
Closest papers by meaning
- IKEA Manuals at Work: 4D Grounding of Assembly Instructions on Internet Videos (NeurIPS 2024 · similarity 0.88)
- Two by Two: Learning Multi-Task Pairwise Objects Assembly for Generalizable Robot Manipulation (CVPR 2025 · similarity 0.88)
- Neural Assembler: Learning to Generate Fine-Grained Robotic Assembly Instructions from Multi-View Images (AAAI 2025 · similarity 0.88)
- Imagine: Image-Guided 3D Part Assembly with Structure Knowledge Graph (AAAI 2025 · similarity 0.88)
- State-Based Disassembly Planning (AAAI 2025 · similarity 0.86)
- Generative 3D Part Assembly via Part-Whole-Hierarchy Message Passing (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.
