Pre-Trained Image Encoder for Generalizable Visual Reinforcement Learning.
Zhecheng YuanZhengrong XueBo YuanXueqian WangYi WuYang GaoHuazhe XuPublished in: CoRR (2022)
Keyphrases
- pre trained
- reinforcement learning
- image features
- input image
- training data
- image segmentation
- high resolution
- feature points
- single image
- visual information
- illumination conditions
- image representation
- pixel level
- image retrieval
- training examples
- image classification
- lighting conditions
- keypoints
- data sets
- face recognition
- learning algorithm