Coarse-to-Fine Q-attention: Efficient Learning for
Vision-based Robotic Manipulation via Discretisation
Stephen James Kentaro Wada Tristan Laidlow Andrew J. Davison
Dyson Robotics Lab, Imperial College London
CVPR 2022, Oral
Code: https://github.com/stepjam/ARM
(Repo contains ARM and C2F-ARM)
Real World Experiments
Real World Experiments
![](https://www.google.com/images/icons/product/drive-32.png)
Take lid off saucepan
Take lid off saucepan
![](https://www.google.com/images/icons/product/drive-32.png)
Fold towel
Fold towel
![](https://www.google.com/images/icons/product/drive-32.png)
Pull toy car
Pull toy car
![](https://www.google.com/images/icons/product/drive-32.png)
Turn on light
Turn on light
![](https://www.google.com/images/icons/product/drive-32.png)
Pull cloth from shelf
Pull cloth from shelf
Videos show the coarse-to-fine Q-attention
![](https://www.google.com/images/icons/product/drive-32.png)
Take Lid Off Saucepan
Take Lid Off Saucepan
![](https://www.google.com/images/icons/product/drive-32.png)
Put Rubbish In Bin
Put Rubbish In Bin