ActiveMimic: Egocentric Video Pretraining with Active Perception
Treats head-camera motion in egocentric human video as a viewpoint action and pretrains active perception plus manipulation from it. Reports matching state-of-the-art models pretrained on robot data.把第一人称人类视频中的头部相机运动视为“视角动作”，同时预训练主动感知与操作能力。报告效果与使用机器人数据预训练的最先进模型相当。
BibTeX
@misc{lin2026activemimic,
title = {{ActiveMimic: Egocentric Video Pretraining with Active Perception}},
author = {Xingyao Lin and Guojin Zhong and Tianyi Lu and Ziyi Ye and Yichen Zhu and Zuxuan Wu and Yu-Gang Jiang},
year = {2026},
eprint = {2606.06194},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2606.06194},
}Robot Learning: RL & Imitation机器人学习：强化学习与模仿学习egocentric-videoactive-perceptionpretrainingcamera-motionreal