(#) NLoS Benchmark

Welcome to the Non-Line-of-Sight Imaging Benchmark!

On this website, we provide reference data and evaluation tools that aim to make existing NLoS reconstruction algorithms more comparable. The website is accompanied by a [paper](paper) published at the [2018 British Machine Vision Conference](http://bmvc2018.org/).

You are free to use the data and tools provided here to benchmark your own non-line-of-sight reconstruction algorithms. If you do, please cite the paper as follows.

Jonathan Klein, Martin Laurenzis, Dominik L. Michels, Matthias B. Hullin, 2018. A Quantitative Platform for Non-Line-of-Sight Imaging Problems. Proceedings of British Machine Vision Conference (BMVC) 2018, Northumbria University, Newcastle, UK, September 3-6, 2018.

The benchmark is published by the [Computational Light Transport / Digital Material Appearance Group](https://light.cs.uni-bonn.de) within the [Institute of Computer Science](https://www.cs.uni-bonn.de) at the [University of Bonn](https://www.uni-bonn.de/).

This website has been developed using funds from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (Grant agreement No. 802192 / ECHO).


**Presentation at the BMVC**


Our benchmark is published at the BMVC 2018 today. Check out our [poster](/paper) during the poster session and have a talk with the authors!

The challenges

                    (##)  Geometry Reconstruction

[  ](/challenges/Geometry)

What is the shape of the hidden object?                
                    (##)  Object Tracking

[  ](/challenges/Tracking)

How does the object move, and what is its orientation?                

                    (##)  Object Classification

[  ](/challenges/Classification)

From a set of known objects, which one is currently hidden in the scene?
                    (##)  Texture Reconstruction

[  ](/challenges/Texture)

How is the hidden object painted?