Track trials, findings, and clinical evidence as it lands.
Search papers, labs, and topics across Lattice.
To bridge the quality gap between low-cost satellite photogrammetry and expensive aerial LiDAR, this work adapts Stable Diffusion 3 to denoise and super-resolve photogrammetric Digital Surface Models (DSMs). The authors prune the text stream and apply a custom patch-wise normalization strategy to transfer natural image diffusion priors to continuous elevation data using multimodal conditioning on DSMs and Pléiades optical imagery. Evaluated on French urban environments, the framework sharply reduces dense urban elevation RMSE from 6.00 m to 3.45 m in-context and from 4.16 m to 2.77 m in held-out Bordeaux.
Repurposing Stable Diffusion 3 for multimodal satellite conditioning cuts 3D elevation error by over 40%, closing the long-standing fidelity gap between cheap orbital photogrammetry and costly aerial LiDAR.
Large-scale Digital Surface Models (DSMs) can be produced cost-effectively from satellite images via stereo-photogrammetry. However, the resulting 3D maps are often contaminated by noise, outliers, and voids. On the other hand, aerial LiDAR provides high-accuracy elevation measurements at a substantially higher cost. In this work, we study diffusion models conditioned both on photogrammetric DSMs and Pléiades imagery to refine vertically co-registered DSMs. We introduce a modified Stable Diffusion 3 architecture with a pruned text stream and a patch-wise normalization strategy, enabling stable training on LiDAR data and transfer from natural images to elevation maps. Experiments in French cities demonstrate that multimodal conditioning improves elevation accuracy, reducing Dense Urban RMSE from 6.00 to 3.45 m in the in-context cities and from 4.16 to 2.77 m in the held-out city of Bordeaux.