1. [Publications](/index.php/publications)
2. Tackling 3D ToF Artifacts Through Learning and the FLAT Dataset
 
 # Tackling 3D ToF Artifacts Through Learning and the FLAT Dataset

  ![](/sites/default/files/styles/wide/public/publications/FLAT_2.jpg?itok=YzZ98soe)

 Scene motion, multiple reflections, and sensor noise introduce artifacts in the depth reconstruction performed by time-of-flight cameras. We propose a two-stage, deep-learning approach to address all of these sources of artifacts simultaneously. We also introduce FLAT, a synthetic dataset of 2000 ToF measurements that capture all of these nonidealities, and can be used to simulate different hardware. Using the Kinect camera as a baseline, we show improved reconstruction errors on simulated and real data, as compared with state-of-the-art methods.



 ## Authors



Qi Guo (SEAS, Harvard University)

[Iuri Frosio](/index.php/person/iuri-frosio)

Orazio Gallo (NVIDIA)

Todd Zickler (SEAS, Harvard University)

[Jan Kautz](/index.php/person/jan-kautz)

 

 

 ## Publication Date



Monday, September 10, 2018

 

 ## Published in



[ECCV 2018](https://eccv2018.org/)

 

 ## Research Area



[Computer Vision](/index.php/research-area/computer-vision)

[Artificial Intelligence and Machine Learning ](/index.php/research-area/machine-learning-artificial-intelligence)

 

 

 ## External Links



[Code and full dataset (>500G!) available here](https://github.com/NVlabs/FLAT)

[Official ECCV paper](http://openaccess.thecvf.com/content_ECCV_2018/papers/Qi_Guo_Tackling_3D_ToF_ECCV_2018_paper.pdf)

 

 

 ## Uploaded Files



[Paper](https://research.nvidia.com/sites/default/files/pubs/2018-09_Tackling-3D-ToF/tof_eccv18_0.pdf "Open file in new window")6.04 MB

[Supplementary material](https://research.nvidia.com/sites/default/files/pubs/2018-09_Tackling-3D-ToF/supplementary.pdf "Open file in new window")7.06 MB