# Learning Navigation Subroutines from Egocentric Videos

arXiv: https://arxiv.org/abs/1905.12612 · HTML: https://mlsanity.com/paper/1905.12612/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 May 2019.

## Built on by (most cited first, 7 of 7)

- [State-Only Imitation Learning for Dexterous Manipulation](https://mlsanity.com/paper/2004.04650/) (2020)
- [Learning to Drive by Watching YouTube Videos: Action-Conditioned Contrastive Policy Pretraining](https://mlsanity.com/paper/2204.02393/) (2022)
- [Zero-shot Imitation Learning from Demonstrations for Legged Robot Visual Navigation](https://mlsanity.com/paper/1909.12971/) (2019)
- [Learning Value Functions from Undirected State-only Experience](https://mlsanity.com/paper/2204.12458/) (2022)
- [InstruGen: Automatic Instruction Generation for Vision-and-Language Navigation Via Large Multimodal Models](https://mlsanity.com/paper/2411.11394/) (2024)
- [One-shot Visual Imitation via Attributed Waypoints and Demonstration Augmentation](https://mlsanity.com/paper/2302.04856/) (2023)
- [Hierarchical Reinforcement Learning for Temporal Pattern Prediction](https://mlsanity.com/paper/2310.05695/) (2023)

## Abstract (arXiv)

Planning at a higher level of abstraction instead of low level torques improves the sample efficiency in reinforcement learning, and computational efficiency in classical planning. We propose a method to learn such hierarchical abstractions, or subroutines from egocentric video data of experts performing tasks. We learn a self-supervised inverse model on small amounts of random interaction data to pseudo-label the expert egocentric videos with agent actions. Visuomotor subroutines are acquired from these pseudo-labeled videos by learning a latent intent-conditioned policy that predicts the inferred pseudo-actions from the corresponding image observations. We demonstrate our proposed approach in context of navigation, and show that we can successfully learn consistent and diverse visuomotor subroutines from passive egocentric videos. We demonstrate the utility of our acquired visuomotor subroutines by using them as is for exploration, and as sub-policies in a hierarchical RL framework for reaching point goals and semantic goals. We also demonstrate behavior of our subroutines in the real world, by deploying them on a real robotic platform. Project website: https://ashishkumar1993.github.io/subroutines/.