# Where to Look Next: Unsupervised Active Visual Exploration on 360deg Input

arXiv: https://arxiv.org/abs/1909.10304 · HTML: https://mlsanity.com/paper/1909.10304/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Sep 2019.

## Built on by (most cited first, 5 of 5)

- [Active Visual Exploration Based on Attention-Map Entropy](https://mlsanity.com/paper/2303.06457/) (2023)
- [Glimpse-Attend-and-Explore: Self-Attention for Active Visual Exploration](https://mlsanity.com/paper/2108.11717/) (2021)
- [AdaGlimpse: Active Visual Exploration with Arbitrary Glimpse Position and Scale](https://mlsanity.com/paper/2404.03482/) (2024)
- [Visual Attention in Imaginative Agents](https://mlsanity.com/paper/2104.00177/) (2021)
- [Symmetry-aware Neural Architecture for Embodied Visual Navigation](https://mlsanity.com/paper/2112.09515/) (2021)

## Abstract (arXiv)

We address the problem of active visual exploration of large 360deg inputs. In our setting an active agent with a limited camera bandwidth explores its 360deg environment by changing its viewing direction at limited discrete time steps. As such, it observes the world as a sequence of narrow field-of-view 'glimpses', deciding for itself where to look next. Our proposed method exceeds previous works' performance by a significant margin without the need for deep reinforcement learning or training separate networks as sidekicks. A key component of our system are the spatial memory maps that make the system aware of the glimpses' orientations (locations in the 360deg image). Further, we stress the advantages of retina-like glimpses when the agent's sensor bandwidth and time-steps are limited. Finally, we use our trained model to do classification of the whole scene using only the information observed in the glimpses.