Time-Contrastive Networks: Self-Supervised Learning from Video

TL;DR

This paper introduces a self-supervised learning method called Time-Contrastive Networks that learns viewpoint-invariant representations from unlabeled videos, enabling robots to imitate human behaviors and poses without explicit supervision.

Contribution

The paper presents a novel self-supervised approach using metric learning to learn robust, viewpoint-invariant representations from unlabeled videos for robotic imitation tasks.

Findings

Robots can imitate human poses directly from learned representations.

The learned representations serve as effective reward functions for reinforcement learning.

The approach works with minimal demonstration data, enabling practical robotic learning.

Abstract

We propose a self-supervised approach for learning representations and robotic behaviors entirely from unlabeled videos recorded from multiple viewpoints, and study how this representation can be used in two robotic imitation settings: imitating object interactions from videos of humans, and imitating human poses. Imitation of human behavior requires a viewpoint-invariant representation that captures the relationships between end-effectors (hands or robot grippers) and the environment, object attributes, and body pose. We train our representations using a metric learning loss, where multiple simultaneous viewpoints of the same observation are attracted in the embedding space, while being repelled from temporal neighbors which are often visually similar but functionally different. In other words, the model simultaneously learns to recognize what is common between different-looking images, and what is different between similar-looking images. This signal causes our model to discover attributes that do not change across viewpoint, but do change across time, while ignoring nuisance variables such as occlusions, motion blur, lighting and background. We demonstrate that this representation can be used by a robot to directly mimic human poses without an explicit correspondence, and that it can be used as a reward function within a reinforcement learning algorithm. While representations are learned from an unlabeled collection of task-related videos, robot behaviors such as pouring are learned by watching a single 3rd-person demonstration by a human. Reward functions obtained by following the human demonstrations under the learned representation enable efficient reinforcement learning that is practical for real-world robotic systems. Video results, open-source code and dataset are available at https://sermanet.github.io/imitate