mlsanityData to 14 Oct 2026

S-EO: A Large-Scale Dataset for Geometry-Aware Shadow Detection in Remote Sensing Applications

arXiv 2504.06920 ↗First posted 9 Apr 2025cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

We introduce the S-EO dataset: a large-scale, high-resolution dataset, designed to advance geometry-aware shadow detection. Collected from diverse public-domain sources, including challenge datasets and government providers such as USGS, our dataset comprises 702 georeferenced tiles across the USA, each covering 500x500 m. Each tile includes multi-date, multi-angle WorldView-3 pansharpened RGB images, panchromatic images, and a ground-truth DSM of the area obtained from LiDAR scans. For each image, we provide a shadow mask derived from geometry and sun position, a vegetation mask based on the NDVI index, and a bundle-adjusted RPC model. With approximately 20,000 images, the S-EO dataset establishes a new public resource for shadow detection in remote sensing imagery and its applications to 3D reconstruction. To demonstrate the dataset's impact, we train and evaluate a shadow detector, showcasing its ability to generalize, even to aerial images. Finally, we extend EO-NeRF - a state-of-the-art NeRF approach for satellite imagery - to leverage our shadow predictions for improved 3D reconstructions.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 20 of its 51 references that are in the map. First posted on arXiv 9 Apr 2025.

Builds on

20 of 51 references.

References · 20

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  3. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  4. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  5. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,546
  6. 2024The Road Less Scheduledcited by 163
  7. 2017Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removalcited by 161
  8. 2018Recalibrating Fully Convolutional Networks with Spatial and Channel 'Squeeze & Excitation' Blockscited by 76
  9. 2022Sat-NeRF: Learning Multi-View Satellite Photogrammetry With Transient Objects and Shadow Modeling Using RPC Camerascited by 59
  10. 2017Direction-aware Spatial Context Features for Shadow Detectioncited by 56
  11. 2018Semantic Stereo for Incidental Satellite Imagescited by 53
  12. 2021Shadow Neural Radiance Fields for Multi-view Satellite Photogrammetrycited by 47
8 more
  1. 2019Revisiting Shadow Detection: A New Benchmark Dataset for Complex Worldcited by 34
  2. 2023SILT: Shadow-aware Iterative Label Tuning for Learning to Detect Shadows from Noisy Labelscited by 19
  3. 2023SparseSat-NeRF: Dense Depth Supervised Neural Radiance Fields for Sparse Satellite Imagescited by 18
  4. 2022ShadowNeuS: Neural SDF Reconstruction by Shadow Ray Supervisioncited by 17
  5. 2024SAT-NGP : Unleashing Neural Graphics Primitives for Fast Relightable Transient-Free 3D reconstruction from Satellite Imagerycited by 14
  6. 2022Towards Learning Neural Representations from Shadowscited by 12
  7. 2023SUNDIAL: 3D Satellite Understanding through Direct, Ambient, and Complex Lighting Decompositioncited by 9
  8. 2021Cumulative Assessment for Urban 3D Modelingcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026Diachronic Stereo Matching for Multi-Date Satellite Imagerycited by 4
  2. 2026ShadowGS: Shadow-Aware 3D Gaussian Splatting for Satellite Imagerycited by 3
  3. 2026SwiftGS: Episodic Priors for Immediate Satellite Surface Recoverycited by 0
  4. 2026deSEO: Physics-Aware Dataset Creation for High-Resolution Satellite Image Shadow Removalcited by 0
  5. 2026After a Decade: Bringing Shadow Removal into the Real World with Agentic Training Datacited by 0