Can Pre-Trained Text-to-Image Models Generate Visual Goals for Reinforcement Learning?

Jialu Gao Kaizhe Hu Guowei Xu Huazhe Xu

Published in: NeurIPS (2023)

Keyphrases

pre trained
reinforcement learning
input image
image content
image segmentation
training data
image retrieval
image representation
single image
image classification
visual information
image features
visual features
training examples
feature points
visual data
statistical model
machine learning
high resolution
lighting conditions
viewpoint
appearance variations
learning algorithm