After clicking “Watch Now” you will be prompted to login or join.
Click “Watch Now” to login or join the NVIDIA Developer Program.
A Reinforcement Learning Approach for Sequential Spatial Transformer Networks
Fatemeh Azimi, DFKI
In this session, we discuss the applications of reinforcement learning (RL) for computer vision applications such as classification. We describe how we combine the idea of spatial transformer networks with RL to improve the classifiers' robustness against noise and clutter.