J. de Curtò i DíAz

March 8, 2021

"Open dataset with astounding images from the Moon"

Hey everybody,

I enclose here a dataset [https://github.com/decurtoidiaz/ce4] with astounding images from the Moon from the mission Chang'E-4 and Chang'E-3 that we released together with the following publication. We were the first to provide PNG images available together with scripts to convert from PDS4 and debayer the gray images. Have fun.

Cycle-consistent Generative Adversarial Networks for Neural Style Transfer using data from Chang’E-4. De Curtò and Duvall. 2020. https://arxiv.org/pdf/2011.15104

If you use the dataset or the scripts to convert from PDS4 to PNG in a publication, please cite the paper below:

    @article{DeCurtoandDuvall20,
      author = "J. {de Curt\'o} and R. Duvall",
      title = "Cycle-consistent Generative Adversarial Networks for Neural Style Transfer using data from Chang’E-4",
      journal = "arXiv:2011.11627",
      year = "2020",
    }


To run the scripts (ce4/utils), you need imagemagick on your system:

$ sudo apt-get install imagemagick

These scripts to convert from PDS4 format to PNG are the first openly available at the time of publication. Link is here:

https://github.com/decurtoidiaz/ce4/tree/main/utils

Check out also the follow-up work on VIO and SLAM:

Vulcan Centaur: towards end-to-end real-time perception in lunar rovers.
De Curtò and Duvall. 2020.
https://arxiv.org/pdf/2011.15104

Best regards,
De Curtò i DíAz.