Motion Tracks: A Unified Representation for Human-Robot Transfer in Few-Shot Imitation Learning

TL;DR

This paper introduces a novel imitation learning approach using motion tracks as a unified action representation, enabling robots to learn tasks from minimal human videos and limited demonstrations with high success rates.

Contribution

The paper proposes Motion Track Policy (MT-pi), a new IL method that uses 2D motion trajectories as actions, bridging human videos and robot demonstrations for efficient task learning.

Findings

Achieves 86.5% success rate across 4 real-world tasks.

Outperforms state-of-the-art IL baselines by 40%.

Generalizes to scenarios only seen in human videos.

Abstract

Teaching robots to autonomously complete everyday tasks remains a challenge. Imitation Learning (IL) is a powerful approach that imbues robots with skills via demonstrations, but is limited by the labor-intensive process of collecting teleoperated robot data. Human videos offer a scalable alternative, but it remains difficult to directly train IL policies from them due to the lack of robot action labels. To address this, we propose to represent actions as short-horizon 2D trajectories on an image. These actions, or motion tracks, capture the predicted direction of motion for either human hands or robot end-effectors. We instantiate an IL policy called Motion Track Policy (MT-pi) which receives image observations and outputs motion tracks as actions. By leveraging this unified, cross-embodiment action space, MT-pi completes tasks with high success given just minutes of human video and limited additional robot demonstrations. At test time, we predict motion tracks from two camera views, recovering 6DoF trajectories via multi-view synthesis. MT-pi achieves an average success rate of 86.5% across 4 real-world tasks, outperforming state-of-the-art IL baselines which do not leverage human data or our action space by 40%, and generalizes to scenarios seen only in human videos. Code and videos are available on our website https://portal-cornell.github.io/motion_track_policy/.