English

A Geometric Perspective on Visual Imitation Learning

Robotics 2020-03-06 v1 Machine Learning

Abstract

We consider the problem of visual imitation learning without human supervision (e.g. kinesthetic teaching or teleoperation), nor access to an interactive reinforcement learning (RL) training environment. We present a geometric perspective to derive solutions to this problem. Specifically, we propose VGS-IL (Visual Geometric Skill Imitation Learning), an end-to-end geometry-parameterized task concept inference method, to infer globally consistent geometric feature association rules from human demonstration video frames. We show that, instead of learning actions from image pixels, learning a geometry-parameterized task concept provides an explainable and invariant representation across demonstrator to imitator under various environmental settings. Moreover, such a task concept representation provides a direct link with geometric vision based controllers (e.g. visual servoing), allowing for efficient mapping of high-level task concepts to low-level robot actions.

Keywords

Cite

@article{arxiv.2003.02768,
  title  = {A Geometric Perspective on Visual Imitation Learning},
  author = {Jun Jin and Laura Petrich and Masood Dehghan and Martin Jagersand},
  journal= {arXiv preprint arXiv:2003.02768},
  year   = {2020}
}

Comments

submitted to IROS 2020

R2 v1 2026-06-23T14:05:25.415Z