Minha Lee
Minha Lee is a researcher affiliated with Korea University who co-authored Pose6DAug, a physically plausible multi-view object swapping framework for robot data augmentation. The framework improves vision-language-action policy generalization to novel objects by leveraging existing successful robot manipulation episodes and synthesizing new demonstrations through 3D-anchored object swaps without requiring additional data collection. She is listed as an author on the Pose6DAug paper alongside researchers from KAIST and RLWRLD.
“The core achievement of this paper is a framework that recycles a robot's past successful actions to teach it how to handle new objects it has never seen before.”
Source→“By fine-tuning Vision-Language-Action (VLA) policies on this augmented data, they achieved a 16.5% relative to the state-of-the-art baseline on novel objects.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.