mlsanityData to 14 Oct 2026

Enhancing Photogrammetric Digital Surface Models with Pretrained Diffusion Models and Multimodal Conditioning

arXiv 2609.31199 ↗First posted 25 Sep 2026eess.IV cs.CVcited by 0

On the map: Diffusion models → 3D, point cloud and scene generation also fits: Inverse problems and image restoration

Abstract (arXiv)

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 Pl'eiades 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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models. Builds on 16 of its 32 references that are in the map. First posted on arXiv 25 Sep 2026.

Builds on

16 of 32 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Marigold-DC: Zero-Shot Monocular Depth Completion with Guided Diffusion18 Dec

Other references · 15

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  4. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  5. 2022Scalable Diffusion Models with Transformerscited by 6,054
  6. 2022Flow Matching for Generative Modelingcited by 5,702
  7. 2023Adding Conditional Control to Text-to-Image Diffusion Modelscited by 5,503
  8. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,181
  9. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,700
  10. 2023Multisample Flow Matching: Straightening Flows with Minibatch Couplingscited by 278
  11. 2023Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion modelscited by 205
  12. 2018Void Filling of Digital Elevation Models with Deep Generative Modelscited by 7
3 more
  1. 2025MESA: Text-Driven Terrain Generation Using Latent Diffusion and Global Copernicus Datacited by 7
  2. 2025A Diffusion-Based Framework for Terrain-Aware Remote Sensing Image Reconstructioncited by 6
  3. 2025Dfilled: Repurposing Edge-Enhancing Diffusion for Guided DSM Void Fillingcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Fast LiDAR Upsampling using Conditional Diffusion Models
  2. 2026Guided Super-Resolution of Digital Elevation Models with Diffusion-Based Image Generators
  3. 2023Single-View Height Estimation with Conditional Diffusion Probabilistic Models
  4. 2025LiDPM: Rethinking Point Diffusion for Lidar Scene Completion
  5. 2023LiDAR Data Synthesis with Denoising Diffusion Probabilistic Models
  6. 2024Scaling Diffusion Models to Real-World 3D LiDAR Scene Completion
  7. 2024Towards Realistic Scene Generation with LiDAR Diffusion Models
  8. 2025PVNet: Point-Voxel Interaction LiDAR Scene Upsampling Via Diffusion Models