Once data is aligned and normalized, they train a unified imitation learning policy that is co-trained on both human and robot data.

So its cotrained rip

Citation

(Kareer et al., 2024)

Kareer, S., Patel, D., Punamiya, R., Mathur, P., Cheng, S., Wang, C., Hoffman, J., & Xu, D. (2024). EgoMimic: Scaling Imitation Learning via Egocentric Video. https://arxiv.org/pdf/2410.24221