# Learning to Visually Navigate in Photorealistic Environments Without any Supervision

arXiv: https://arxiv.org/abs/2004.04954 · HTML: https://mlsanity.com/paper/2004.04954/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Apr 2020.

## Built on by (most cited first, 6 of 6)

- [A Survey of Embodied AI: From Simulators to Research Tasks](https://mlsanity.com/paper/2103.04918/) (2021)
- [SOON: Scenario Oriented Object Navigation with Graph-based Exploration](https://mlsanity.com/paper/2103.17138/) (2021)
- [Memory-Augmented Reinforcement Learning for Image-Goal Navigation](https://mlsanity.com/paper/2101.05181/) (2021)
- [Active Audio-Visual Separation of Dynamic Sound Sources](https://mlsanity.com/paper/2202.00850/) (2022)
- [Symmetry-aware Neural Architecture for Embodied Visual Navigation](https://mlsanity.com/paper/2112.09515/) (2021)
- [FOD-S2R: A FOD Dataset for Sim2Real Transfer Learning based Object Detection](https://mlsanity.com/paper/2512.01315/) (2025)

## Abstract (arXiv)

Learning to navigate in a realistic setting where an agent must rely solely on visual inputs is a challenging task, in part because the lack of position information makes it difficult to provide supervision during training. In this paper, we introduce a novel approach for learning to navigate from image inputs without external supervision or reward. Our approach consists of three stages: learning a good representation of first-person views, then learning to explore using memory, and finally learning to navigate by setting its own goals. The model is trained with intrinsic rewards only so that it can be applied to any environment with image observations. We show the benefits of our approach by training an agent to navigate challenging photo-realistic environments from the Gibson dataset with RGB inputs only.