Research map / Detection and segmentation / Open-set and few-shot detection
Grasp Proposal Networks: An End-to-End Solution for Visual Learning of Robotic Grasps
Chaozheng Wu, Jian Chen, Qiaoyu Cao, Jianchi Zhang, Yunxin Tai, Lin Sun, Kui Jia
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 3D grid anchors and retain anchor coordinates in proposal features (ablation) | GPNet · Test set | rule-based success rate@10%: 0.372 -> 0.933 | Improved |
| Increase grid resolution from r=3 to r=10 at b=22cm (ablation) | GPNet · Test set | simulation success rate@10%: 0.644 -> 0.900 | Improved |
| Spread anchors by expanding proposal space from 10cm to 30cm at r=10 (ablation) | GPNet · Test set | coverage rate@10%: 0.055 -> 0.073 | Improved |
| Expand proposal-space length from 10cm to 30cm at r=10 (ablation) | GPNet · Test set | simulation success rate@10%: 0.494 -> 0.833 | Improved |
| Expand proposal-space length from 22cm to 30cm at r=10 (ablation) | GPNet · Test set | rule-based success rate@10%: 0.933 -> 0.900 | Hurt |
| Increase annotations per object from 10K to 100K (ablation) | GPNet · Test set | simulation success rate@10%: 0.650 -> 0.900 | Improved |
| Increase training-set ratio from 1/4 to 1 (ablation) | GPNet · Test set | simulation success rate@10%: 0.522 -> 0.900 | Improved |
| Use GPNet rather than 6-DOF GraspNet with refinement | 6-DOF grasping · Test set | simulation success rate@10%: 0.800 -> 0.900 | Improved |
| Use GPNet rather than 6-DOF GraspNet | Real robot grasping · 20-object real robot test | success rate: 73% -> 85% | Improved |
Closest papers by meaning
- Generalizing 6-DoF Grasp Detection via Domain Prior Knowledge (CVPR 2024 · similarity 0.92)
- GraspNet-1Billion: A Large-Scale Benchmark for General Object Grasping (CVPR 2020 · similarity 0.92)
- UniGraspTransformer: Simplified Policy Distillation for Scalable Dexterous Robotic Grasping (CVPR 2025 · similarity 0.88)
- ZeroGrasp: Zero-Shot Shape Reconstruction Enabled Robotic Grasping (CVPR 2025 · similarity 0.88)
- Active Perception for Grasp Detection via Neural Graspness Field (NeurIPS 2024 · similarity 0.88)
- Grasp2Grasp: Vision-Based Dexterous Grasp Translation via Schrödinger Bridges (NeurIPS 2025 · similarity 0.88)
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.
