# What You Can Learn by Staring at a Blank Wall

arXiv: https://arxiv.org/abs/2108.13027 · HTML: https://mlsanity.com/paper/2108.13027/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 46 references that are in the map.
- First posted on arXiv 30 Aug 2021.

## Builds on (most cited first, 13 of 13)

- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Seeing Around Street Corners: Non-Line-of-Sight Detection and Tracking In-the-Wild Using Doppler Radar](https://mlsanity.com/paper/1912.06613/) (2019)
- [Steady-state Non-Line-of-Sight Imaging](https://mlsanity.com/paper/1811.09910/) (2018)
- [Computational Mirrors: Blind Inverse Light Transport by Deep Matrix Factorization](https://mlsanity.com/paper/1912.02314/) (2019)
- [Keyhole Imaging: Non-Line-of-Sight Imaging and Tracking of Moving Objects Along a Single Optical Path](https://mlsanity.com/paper/1912.06727/) (2019)
- [Seeing Around Corners with Edge-Resolved Transient Imaging](https://mlsanity.com/paper/2002.07118/) (2020)
- [Two-Dimensional Non-Line-of-Sight Scene Estimation from a Single Edge Occluder](https://mlsanity.com/paper/2006.09241/) (2020)
- [Non-line-of-sight tracking of people at long range](https://mlsanity.com/paper/1703.02124/) (2017)
- [Optically lightweight tracking of objects around a corner](https://mlsanity.com/paper/1606.01873/) (2016)
- [Neural network identification of people hidden from view with a single-pixel, single-photon detector](https://mlsanity.com/paper/1709.07244/) (2017)
- [Flash Photography for Data-Driven Hidden Scene Recovery](https://mlsanity.com/paper/1810.11710/) (2018)
- [Adaptive Lighting for Data-Driven Non-Line-of-Sight 3D Localization and Object Identification](https://mlsanity.com/paper/1905.11595/) (2019)
- [Non-line-of-sight Imaging with Partial Occluders and Surface Normals](https://mlsanity.com/paper/1711.07134/) (2017)

## Built on by (most cited first, 10 of 10)

- [Role of Transients in Two-Bounce Non-Line-of-Sight Imaging](https://mlsanity.com/paper/2304.01308/) (2023)
- [Seeing the World through Your Eyes](https://mlsanity.com/paper/2306.09348/) (2023)
- [Totems: Physical Objects for Verifying Visual Integrity](https://mlsanity.com/paper/2209.13032/) (2022)
- [Can Shadows Reveal Biometric Information?](https://mlsanity.com/paper/2209.10077/) (2022)
- [Propagate And Calibrate: Real-time Passive Non-line-of-sight Tracking](https://mlsanity.com/paper/2303.11791/) (2023)
- [Domain Reduction Strategy for Non Line of Sight Imaging](https://mlsanity.com/paper/2308.10269/) (2023)
- [Invisible-to-Visible: Privacy-Aware Human Instance Segmentation using Airborne Ultrasound via Collaborative Learning Variational Autoencoder](https://mlsanity.com/paper/2204.07280/) (2022)
- [PathFinder: Attention-Driven Dynamic Non-Line-of-Sight Tracking with a Mobile Robot](https://mlsanity.com/paper/2404.05024/) (2024)
- [Wallcamera: Reinventing the Wheel?](https://mlsanity.com/paper/2407.16015/) (2024)
- [Soft Shadow Diffusion (SSD): Physics-inspired Learning for 3D Computational Periscopy](https://mlsanity.com/paper/2601.12257/) (2026)

## Abstract (arXiv)

We present a passive non-line-of-sight method that infers the number of people or activity of a person from the observation of a blank wall in an unknown room. Our technique analyzes complex imperceptible changes in indirect illumination in a video of the wall to reveal a signal that is correlated with motion in the hidden part of a scene. We use this signal to classify between zero, one, or two moving people, or the activity of a person in the hidden scene. We train two convolutional neural networks using data collected from 20 different scenes, and achieve an accuracy of ≈94% for both tasks in unseen test environments and real-time online settings. Unlike other passive non-line-of-sight methods, the technique does not rely on known occluders or controllable light sources, and generalizes to unknown rooms with no re-calibration. We analyze the generalization and robustness of our method with both real and synthetic data, and study the effect of the scene parameters on the signal quality.