Pick and Place on Pick and Place Tomato Soup Can Single Object Level 2
55Success RateCyberDemo
Evaluation Results
| Method | Links | |
|---|---|---|
| CyberDemoPre-training=Augmented simulation demonstrations2024.02 | 55 | |
| MVPBackbone=Vision Transformer, Pre-training=Masked Autoencoder2024.02 | 35 | |
| R3MBackbone=ResNet-50, Pre-training=Time-contrastive, video-language alignment (Ego4D)2024.02 | 15 | |
| PVRBackbone=ResNet-50, Pre-training=MoCo-v2 (ImageNet)2024.02 | 0 |