Challenging planetary DTM dataset

Please cite the associated paper when using the dataset : Brun, Loïs, Doute, Sylvain, Bernard-Salas, Jeronimo, and Paiement, Adeline. “3D registration of remote sensing data for planetary exploration.” In CVPR 2026 Workshop on 3D Geometry Generation for Scientific Computing, 2026.

These datasets are representative of the properties and artifacts that may be encountered in the context of Digital Elevation/Terrain Models (DEM/DTM) in planetary remote sensing. The spatial resolution can vary widely, from submeter to hundred of meters, additionally resulting in uneven sampling and sometimes partial overlap. Martian data also share challenges with Earth remote sensing, while being affected by higher uncertainties, thus allowing to assess the limit performance of compared methods.

These datasets are available in 5 subfolders in clear, explicit and understandable naming conventions.

These datasets are a product of the PhD of Loïs Brun and is presented in the associated articles. This work was carried out as part of a PhD thesis co-funded by the University of Toulon, IPAG, and ACRI-ST.