Reproduce “Enhancing Photogrammetric Digital Surface Models with Pretrained Diffusion Models and Multimodal Conditioning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Antoine Lorentz, Stéphane May, Valentine Bellet, Dawa Derksen, Bastien Nespoulous
The authors refine calibrated satellite stereo DSMs by adapting Stable Diffusion 3 to combine the DSM with Pléiades RGB imagery, using LiDAR for training but not inference. In French CARS/Pléiades tests, Dense Urban RMSE falls from 6.00 to 3.45 metres in-context and from 4.16 to 2.77 metres in held-out Bordeaux. Results assume co-registration and compatible surfaces.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.