mlsanityData to 14 Oct 2026

DoF-Gaussian: Controllable Depth-of-Field for 3D Gaussian Splatting

arXiv 2503.00746 ↗First posted 2 Mar 2025cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Recent advances in 3D Gaussian Splatting (3D-GS) have shown remarkable success in representing 3D scenes and generating high-quality, novel views in real-time. However, 3D-GS and its variants assume that input images are captured based on pinhole imaging and are fully in focus. This assumption limits their applicability, as real-world images often feature shallow depth-of-field (DoF). In this paper, we introduce DoF-Gaussian, a controllable depth-of-field method for 3D-GS. We develop a lens-based imaging model based on geometric optics principles to control DoF effects. To ensure accurate scene geometry, we incorporate depth priors adjusted per scene, and we apply defocus-to-focus adaptation to minimize the gap in the circle of confusion. We also introduce a synthetic dataset to assess refocusing capabilities and the model's ability to learn precise lens parameters. Our framework is customizable and supports various interactive applications. Extensive experiments confirm the effectiveness of our method. Our project is available at https://dof-gaussian.github.io.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 42 of its 61 references that are in the map. First posted on arXiv 2 Mar 2025.

Builds on

42 of 61 references.

References · 42

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  2. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  3. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,495
  4. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,274
  5. 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,232
  6. 2020Nerfies: Deformable Neural Radiance Fieldscited by 1,198
  7. 20234D Gaussian Splatting for Real-Time Dynamic Scene Renderingcited by 1,157
  8. 20242D Gaussian Splatting for Geometrically Accurate Radiance Fieldscited by 1,076
  9. 2023Mip-Splatting: Alias-free 3D Gaussian Splattingcited by 667
  10. 2024Depth Pro: Sharp Monocular Metric Depth in Less Than a Secondcited by 404
  11. 2021NeRF in the Dark: High Dynamic Range View Synthesis from Noisy Raw Imagescited by 336
  12. 2024DepthSplat: Connecting Gaussian Splatting and Depthcited by 286
30 more
  1. 2024DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalizationcited by 256
  2. 2021NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereocited by 192
  3. 2021Deblur-NeRF: Neural Radiance Fields from Blurry Imagescited by 179
  4. 2023Depth-Regularized Optimization for 3D Gaussian Splatting in Few-Shot Imagescited by 173
  5. 20243D Gaussian Ray Tracing: Fast Tracing of Particle Scenescited by 139
  6. 2019Learning the Depths of Moving People by Watching Frozen Peoplecited by 120
  7. 2024RaDe-GS: Rasterizing Depth in Gaussian Splattingcited by 111
  8. 2024MVSGaussian: Fast Generalizable Gaussian Splatting Reconstruction from Multi-View Stereocited by 87
  9. 2022BAD-NeRF: Bundle Adjusted Deblur Neural Radiance Fieldscited by 80
  10. 2018Synthetic Depth-of-Field with a Single-Camera Mobile Phonecited by 71
  11. 2022DP-NeRF: Deblurred Neural Radiance Field with Physical Scene Priorscited by 70
  12. 2024Deblurring 3D Gaussian Splattingcited by 66
  13. 2024BAD-Gaussians: Bundle Adjusted Deblur Gaussian Splattingcited by 61
  14. 2020Rendering Natural Camera Bokeh Effect with Deep Learningcited by 57
  15. 2022BokehMe: When Neural Rendering Meets Classical Renderingcited by 48
  16. 2024BAGS: Blur Agnostic Gaussian Splatting through Multi-Scale Kernel Modelingcited by 42
  17. 2019Synthetic Defocus and Look-Ahead Autofocus for Casual Videographycited by 28
  18. 2024Lighting Every Darkness with 3DGS: Fast Training and Real-Time Rendering for HDR View Synthesiscited by 27
  19. 2020AIM 2020 Challenge on Rendering Realistic Bokehcited by 26
  20. 2022DoF-NeRF: Depth-of-Field Meets Neural Radiance Fieldscited by 26
  21. 2019SteReFo: Efficient Image Refocusing with Stereo Visioncited by 25
  22. 2020BGGAN: Bokeh-Glass Generative Adversarial Network for Rendering Realistic Bokehcited by 19
  23. 2023Dr.Bokeh: DiffeRentiable Occlusion-aware Bokeh Renderingcited by 19
  24. 2023Make-It-4D: Synthesizing a Consistent Long-Term Dynamic Scene Video from a Single Imagecited by 19
  25. 2021Stacked Deep Multi-Scale Hierarchical Network for Fast Bokeh Effect Rendering from a Single Imagecited by 18
  26. 2024DyBluRF: Dynamic Neural Radiance Fields from Blurry Monocular Videocited by 17
  27. 2024Robust Gaussian Splattingcited by 16
  28. 2024Cinematic Gaussians: Real-Time HDR Radiance Fields with Depth of Fieldcited by 12
  29. 2022NeRFocus: Neural Radiance Field for 3D Synthetic Defocuscited by 8
  30. 2024Dynamic Neural Radiance Field From Defocused Monocular Videocited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 3.

6 papers.

  1. 2025Generative Refocusing: Flexible Defocus Control from a Single Imagecited by 4
  2. 2025DET-GS: Depth- and Edge-Aware Regularization for High-Fidelity 3D Gaussian Splattingcited by 2
  3. 2026Benchmarking PhD-Level Coding in 3D Geometric Computer Visioncited by 2
  4. 2026Unblur-SLAM: Dense Neural SLAM for Blurry Inputscited by 1
  5. 2025Every Camera Effect, Every Time, All at Once: 4D Gaussian Ray Tracing for Physics-based Camera Effect Data Generationcited by 0
  6. 2026Practical High-Fidelity Novel-View Synthesis of Mounted Lepidopteracited by 0