mlsanityData to 7 Oct 2026

OutCast: Outdoor Single-image Relighting with Cast Shadows

arXiv 2204.09341 ↗First posted 20 Apr 2022cs.GR cs.CV cs.LGcited by 30

Not on the map. Why?

Abstract (arXiv)

We propose a relighting method for outdoor images. Our method mainly focuses on predicting cast shadows in arbitrary novel lighting directions from a single image while also accounting for shading and global effects such the sun light color and clouds. Previous solutions for this problem rely on reconstructing occluder geometry, e.g. using multi-view stereo, which requires many images of the scene. Instead, in this work we make use of a noisy off-the-shelf single-image depth map estimation as a source of geometry. Whilst this can be a good guide for some lighting effects, the resulting depth map quality is insufficient for directly ray-tracing the shadows. Addressing this, we propose a learned image space ray-marching layer that converts the approximate depth map into a deep 3D representation that is fused into occlusion queries using a learned traversal. Our proposed method achieves, for the first time, state-of-the-art relighting results, with only a single image as input. For supplementary material visit our project page at: https://dgriffiths.uk/outcast.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Builds on 46 of its 94 references that are in the map. First posted on arXiv 20 Apr 2022.

Builds on

46 of 94 references.

References · 46

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  4. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  5. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,574
  6. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,484
  7. 2017High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANscited by 1,176
  8. 2016Least Squares Generative Adversarial Networkscited by 1,000
  9. 2017Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networkscited by 707
  10. 2020NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesiscited by 493
  11. 2014Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecturecited by 483
  12. 2020NeRD: Neural Reflectance Decomposition from Image Collectionscited by 446
34 more
  1. 2021ARKitScenes: A Diverse Real-World Dataset For 3D Indoor Scene Understanding Using Mobile RGB-D Datacited by 442
  2. 2020Learning to Recover 3D Scene Shape from a Single Imagecited by 214
  3. 2019Inverse Rendering for Complex Indoor Scenes: Shape, Spatially-Varying Lighting and SVBRDF from a Single Imagecited by 202
  4. 2020Neural Reflectance Fields for Appearance Acquisitioncited by 196
  5. 2018Escaping Plato's Cave: 3D Shape From Adversarial Renderingcited by 170
  6. 2017Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removalcited by 160
  7. 2019Single Image Portrait Relightingcited by 152
  8. 2021NeRFactor: Neural Factorization of Shape and Reflectance Under an Unknown Illuminationcited by 135
  9. 2020Epipolar Transformerscited by 117
  10. 2020Deep Reflectance Volumes: Relightable Reconstructions from Multi-View Photometric Imagescited by 98
  11. 2019Learning Physics-guided Face Relighting under Directional Lightcited by 90
  12. 2019Deep Parametric Indoor Lighting Estimationcited by 82
  13. 2019Rendering Synthetic Objects into Legacy Photographscited by 72
  14. 2017Neural Face Editing with Intrinsic Image Disentanglingcited by 70
  15. 2019Deep Sky Modeling for Single Image Outdoor Lighting Estimationcited by 66
  16. 2019DeepLight: Learning Illumination for Unconstrained Mobile Mixed Realitycited by 63
  17. 2019Inverse Path Tracing for Joint Material and Lighting Estimationcited by 62
  18. 2016Deep Outdoor Illumination Estimationcited by 53
  19. 2021Free-viewpoint Indoor Neural Relighting from Multi-view Stereocited by 50
  20. 2021Self-supervised Outdoor Scene Relightingcited by 48
  21. 2016Deep Shading: Convolutional Neural Networks for Screen-Space Shadingcited by 46
  22. 2019E-LPIPS: Robust Perceptual Image Similarity via Random Transformation Ensemblescited by 44
  23. 2019All-Weather Deep Outdoor Lighting Estimationcited by 41
  24. 2020Learning to Factorize and Relight a Citycited by 36
  25. 2021Single-image Full-body Human Relightingcited by 30
  26. 2020SSN: Soft Shadow Network for Image Compositingcited by 26
  27. 2020Light Stage Super-Resolution: Continuous High-Frequency Relightingcited by 20
  28. 2019Real-Time Global Illumination Decomposition of Videoscited by 18
  29. 2019Relighting Humans: Occlusion-Aware Inverse Rendering for Full-Body Human Imagescited by 16
  30. 2013Shadow Detection: A Survey and Comparative Evaluation of Recent Methodscited by 10
  31. 2019Shadow Transfer: Single Image Relighting For Urban Road Scenescited by 5
  32. 2018MONET: Multiview Semi-supervised Keypoint Detection via Epipolar Divergencecited by 4
  33. 2018Epipolar Geometry based Learning of Multi-view Depth and Ego-Motion from Monocular Sequencescited by 3
  34. 2019X-Section: Cross-Section Prediction for Enhanced RGBD Fusioncited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 2; 2024: 7; 2025: 13; 2026: 7.

30 papers.

Sort:

By area:

  1. 2024DiLightNet: Fine-grained Lighting Control for Diffusion-based Image Generationcited by 93
  2. 2024Neural Gaffer: Relighting Any Object via Diffusioncited by 91
  3. 2022Learning-based Inverse Rendering of Complex Indoor Scenes with Differentiable Monte Carlo Raytracingcited by 78
  4. 2024LightIt: Illumination Modeling and Control for Diffusion Modelscited by 56
  5. 2025DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Modelscited by 55
  6. 2023Relighting Neural Radiance Fields with Shadow and Highlight Hintscited by 41
  7. 2025UniRelight: Learning Joint Decomposition and Synthesis for Video Relightingcited by 38
  8. 2024A Diffusion Approach to Radiance Field Relighting using Multi-Illumination Synthesiscited by 37
  9. 2025LightLab: Controlling Light Sources in Images with Diffusion Modelscited by 36
  10. 2023EverLight: Indoor-Outdoor Editable HDR Lighting Estimationcited by 32
  11. 2025Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Datasetcited by 28
  12. 2024GenLit: Reformulating Single-Image Relighting as Video Generationcited by 20
18 more
  1. 2025IntrinsicEdit: Precise generative image manipulation in intrinsic spacecited by 17
  2. 2025Ouroboros: Single-step Diffusion Models for Cycle-consistent Forward and Inverse Renderingcited by 16
  3. 2025Physically Controllable Relighting of Photographscited by 14
  4. 2025GaSLight: Gaussian Splats for Spatially-Varying Lighting in HDRcited by 11
  5. 2025GaRe: Relightable 3D Gaussian Splatting for Outdoor Scenes from Unconstrained Photo Collectionscited by 11
  6. 2025MV-CoLight: Efficient Object Compositing with Consistent Lighting and Shadow Generationcited by 8
  7. 2025PractiLight: Practical Light Control Using Foundational Diffusion Modelscited by 7
  8. 2026LuxRemix: Lighting Decomposition and Remixing for Indoor Scenescited by 7
  9. 2025MatSwap: Light-aware material transfers in imagescited by 6
  10. 2026PI-Light: Physics-Inspired Diffusion for Full-Image Relightingcited by 5
  11. 2024All-frequency Full-body Human Image Relightingcited by 3
  12. 2024A General Albedo Recovery Approach for Aerial Photogrammetric Images through Inverse Renderingcited by 2
  13. 2026TokenLight: Precise Lighting Control in Images using Attribute Tokenscited by 1
  14. 2025Towards Physically-Based Sky-Modeling For Image Based Lightingcited by 0
  15. 2026Joint Shadow Generation and Relighting via Light-Geometry Interaction Mapscited by 0
  16. 2026SRUG: Shadow-Guided Relightable Urban Scene with Generation Modelcited by 0
  17. 2026BRDFusion: Physics Meets Generation for Urban Scene Inverse Renderingcited by 0
  18. 2026RelightFormer: Feed-forward Generative Transformer for Multiview Object Relightingcited by 0