Robotic Grasp Detection on Cornell Grasp Dataset (Object-wise)
96.1AccuracyOurs
Evaluation Results
| Method | Links | |
|---|---|---|
| Ourspolicy=Top-12018.02 | 96.1 | |
| Ourspolicy=center2018.02 | 96.1 | |
| Guo et al.2018.02 | 89.9 | |
| Multi-Modal Grasp PredictorBackbone=ResNet-50x2, Activation=ReLU, ReLU, ReLU, Input=RGB-D2016.11 | 88.96 | |
| Uni-modal Grasp PredictorBackbone=ResNet-50, Activation=tanh, ReLU, Input=RGD2016.11 | 88.4 | |
| Uni-modal Grasp PredictorBackbone=ResNet-50, Activation=ReLU, ReLU, Input=RGB2016.11 | 87.72 | |
| STEM-CaRFs (Selective Grasp)2016.11 | 87.5 | |
| AlexNet, MultiGrasp2016.11 | 87.1 | |
| Kumra et al.2018.02 | 86.5 | |
| Uni-modal Grasp PredictorBackbone=ResNet-50, Classifier=SVM, Input=RGB, Note=Baseline2016.11 | 84.47 | |
| Multi-Modal Grasp PredictorBackbone=ResNet-50x2, Classifier=linear SVM, Input=RGB-D2016.11 | 84.47 | |
| SAE, struct. reg. two-stage2016.11 | 75.6 | |
| Fast Search2016.11 | 58.3 | |
| Chance2016.11 | 6.7 |