Video2Reward uses keypoint trajectories extracted from videos to let a language model generate and iteratively refine reward functions, improving legged robot behavior learning over text-only methods.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learning
Video2Reward uses keypoint trajectories extracted from videos to let a language model generate and iteratively refine reward functions, improving legged robot behavior learning over text-only methods.