Cornell-Box Dataset

Download

The CornellBox Dataset can be downloaded from this URL

https://viscom.datasets.uni-ulm.de/radu/dataset.zip

Dataset

The dataset contains correlation measurements, ToF depth images and ground truth depth images in .hdr format.

The script simulate_noise_on_correlations.py can be used to simulate shot noise on the correlation images using the default arguments.

Citing this work

If you use this data in your work, please kindly cite the following paper:

@InProceedings{schelling2022radu,
author = {Schelling, Michael and Hermosilla, Pedro and Ropinski, Timo},
title = {{RADU} - Ray-Aligned Depth Update Convolutions for {ToF} Data Denoising},
booktitle = {Conference on Computer Vision and Patter Recognition (CVPR)},
year = {2022}
}

References

The data was generated using the transient renderer of Jarabo et al. [1].

[1] Jarabo, A., Marco, J., Muñoz, A., Buisan, R., Jarosz, W., Gutierrez, D.: "A framework for transient rendering" ACM Transactions on Graphics, SIGGRAPH ASIA, (2014).