Research map / Point clouds / Shape and surface completion
TOSC: Task-Oriented Shape Completion for Open-World Dexterous Grasp Generation from Partial Point Clouds
Weishang Wu, Yifei Shi, Zhiping Cai
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 task-oriented candidate generation (ControlNet + 3D generator) before selection (ablation) | TOSC · OakInk-PartialPC grasping | LLM score: 71.7 -> 88.3 | Improved |
| add DAE plausibility selection and restoration (ablation) | TOSC · OakInk-PartialPC grasping | LLM score: 75.0 -> 88.3 | Improved |
| use task-oriented instead of generic shape completion (ablation) | TOSC · OakInk-PartialPC grasping | LLM score: 66.7 -> 88.3 | Improved |
| add token masking to the 3D DAE (ablation) | TOSC · OakInk-PartialPC grasping | Grasp Displacement mean (cm): 3.21 -> 3.11 | Improved |
| use gradient guidance instead of direct loss-based optimization in FlowGrasp (ablation) | FlowGrasp · OakInk-PartialPC grasping | Grasp Displacement mean (cm): 3.43 -> 3.11 | Improved |
| add DAE selection and restoration (ablation) | TOSC · OakInk-PartialPC grasping | Penetration volume (cm3): 7.96 -> 6.87 | Mixed |
| task-oriented completion plus FlowGrasp vs best prior grasp method | TOSC · OakInk-PartialPC grasping | Penetration volume (cm3): 6.52 (SceneDiffuser) vs 6.87 | No clear effect |
| use TOSC completion vs generic completion methods | TOSC · OakInk-PartialPC shape completion | CD-l2 x 10^-4: 3.71 -> 1.66 | Improved |
| rely on pretrained foundation models for generation | TOSC · Novel categories grasping | P-FID: 46.42 -> 42.97 | Improved |
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.
- PointAttn — PointAttN: You Only Need Attention for Point Cloud Completion (AAAI 2024) · better on 1 of 1 shared result
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
- Contact Map Transfer with Conditional Diffusion Model for Generalizable Dexterous Grasp Generation (NeurIPS 2025 · similarity 0.89)
- CADGrasp: Learning Contact and Collision Aware General Dexterous Grasping in Cluttered Scenes (NeurIPS 2025 · similarity 0.88)
- ZeroGrasp: Zero-Shot Shape Reconstruction Enabled Robotic Grasping (CVPR 2025 · similarity 0.87)
- DexGrasp Anything: Towards Universal Robotic Dexterous Grasping with Physics Awareness (CVPR 2025 · similarity 0.86)
- Single-View Scene Point Cloud Human Grasp Generation (CVPR 2024 · similarity 0.86)
- ManiVideo: Generating Hand-Object Manipulation Video with Dexterous and Generalizable Grasping (CVPR 2025 · 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.
