About this document
R+X: Robot Skill Learning from Videos by marilynlxy is a document available to read on EtoBox.
R+X is a framework that enables robots to learn skills from long, unlabelled videos of humans performing everyday tasks by retrieving relevant video clips based on language commands and executing the learned behaviors through in-context imitation learning. This approach eliminates the need for manual annotation and training, allowing robots to immediately perform tasks without prior training on the retrieved videos. R+X leverages Vision Language Models for both retrieval and execution, demonstrating superio
- Author
- marilynlxy
- Language
- EN