One of the central challenges preventing robots from acquiring complex manipulation skills is the prohibitive cost of collecting large-scale robot demonstrations. In contrast, humans are able to learn efficiently by watching others interact with their environment. To bridge this gap, we introduce semantic action flow as a core intermediate representation capturing the essential spatio-temporal manipulator-object interactions, invariant to superficial visual differences. We present ViSA-Flow, a framework that learns this representation self-supervised from unlabeled large-scale video data. First, a generative model is pre-trained on semantic action flows automatically extracted from large-scale human-object interaction video data, learning a robust prior over manipulation structure. Second, this prior is efficiently adapted to a target robot by fine-tuning on a small set of robot demonstrations processed through the same semantic abstraction pipeline. We demonstrate through extensive experiments on the CALVIN benchmark and real-world tasks that ViSA-Flow achieves state-of-the-art performance, particularly in low-data regimes, outperforming prior methods by effectively transferring knowledge from human video observation to robotic execution. Videos are available atthis https URL.
View on arXiv@article{chen2025_2505.01288,
title={ ViSA-Flow: Accelerating Robot Skill Learning via Large-Scale Video Semantic Action Flow },
author={ Changhe Chen and Quantao Yang and Xiaohao Xu and Nima Fazeli and Olov Andersson },
journal={arXiv preprint arXiv:2505.01288},
year={ 2025 }
}