mlsanityData to 14 Oct 2026

Self-Evolving Depth-Supervised 3D Gaussian Splatting from Rendered Stereo Pairs

arXiv 2409.07456 ↗First posted 11 Sep 2024cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

3D Gaussian Splatting (GS) significantly struggles to accurately represent the underlying 3D scene geometry, resulting in inaccuracies and floating artifacts when rendering depth maps. In this paper, we address this limitation, undertaking a comprehensive analysis of the integration of depth priors throughout the optimization process of Gaussian primitives, and present a novel strategy for this purpose. This latter dynamically exploits depth cues from a readily available stereo network, processing virtual stereo pairs rendered by the GS model itself during training and achieving consistent self-improvement of the scene representation. Experimental results on three popular datasets, breaking ground as the first to assess depth accuracy for these models, validate our findings.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 53 of its 78 references that are in the map. First posted on arXiv 11 Sep 2024.

Builds on

53 of 78 references.

References · 53

  1. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  2. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
  3. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,032
  4. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,964
  5. 2021Vision Transformers for Dense Predictioncited by 1,803
  6. 2021Plenoxels: Radiance Fields without Neural Networkscited by 1,613
  7. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,495
  8. 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,232
  9. 2021Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstructioncited by 964
  10. 2020Neural Sparse Voxel Fieldscited by 959
  11. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
  12. 2020NeRF++: Analyzing and Improving Neural Radiance Fieldscited by 861
38 more
  1. 2018MVSNet: Depth Inference for Unstructured Multi-view Stereocited by 788
  2. 2023ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenescited by 764
  3. 2021Depth-supervised NeRF: Fewer Views and Faster Training for Freecited by 735
  4. 2021KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPscited by 673
  5. 2023Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fieldscited by 671
  6. 2023ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depthcited by 657
  7. 2021RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputscited by 615
  8. 2021Ref-NeRF: Structured View-Dependent Appearance for Neural Radiance Fieldscited by 575
  9. 2021FastNeRF: High-Fidelity Neural Rendering at 200FPScited by 511
  10. 2022Point-NeRF: Point-based Neural Radiance Fieldscited by 503
  11. 2019BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networkscited by 490
  12. 2022MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstructioncited by 470
  13. 2021Dynamic View Synthesis from Dynamic Monocular Videocited by 420
  14. 2021RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matchingcited by 353
  15. 2021Dense Depth Priors for Neural Radiance Fields from Sparse Input Viewscited by 302
  16. 2019Differentiable Surface Splatting for Point-based Geometry Processingcited by 239
  17. 2020Free View Synthesiscited by 231
  18. 2022Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlationcited by 197
  19. 2021InfoNeRF: Ray Entropy Minimization for Few-Shot Neural Volume Renderingcited by 182
  20. 2023Depth-Regularized Optimization for 3D Gaussian Splatting in Few-Shot Imagescited by 173
  21. 2020Stable View Synthesiscited by 165
  22. 2022SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Imagecited by 165
  23. 2022NeRF: Neural Radiance Field in 3D Vision: A Comprehensive Review (Updated Post-Gaussian Splatting)cited by 158
  24. 2021NeRFReN: Neural Radiance Fields with Reflectionscited by 122
  25. 2022EfficientNeRF: Efficient Neural Radiance Fieldscited by 111
  26. 2021T"oRF: Time-of-Flight Radiance Fields for Dynamic Scene View Synthesiscited by 88
  27. 2022Differentiable Point-Based Radiance Fields for Efficient View Synthesiscited by 86
  28. 2020Worldsheet: Wrapping the World in a 3D Sheet for View Synthesis from a Single Imagecited by 77
  29. 2020Monocular Depth Estimation Based On Deep Learning: An Overviewcited by 68
  30. 2022Multiview Stereo with Cascaded Epipolar RAFTcited by 51
  31. 2023NeRF-Supervised Deep Stereocited by 46
  32. 2020Neural Volume Rendering: NeRF And Beyondcited by 44
  33. 2019Monocular Depth Estimation: A Surveycited by 38
  34. 2023BeyondPixels: A Comprehensive Review of the Evolution of Neural Radiance Fieldscited by 27
  35. 2024A Survey on Deep Stereo Matching in the Twentiescited by 23
  36. 2023Revisiting Depth Completion from a Stereo Matching Perspective for Cross-domain Generalizationcited by 18
  37. 2023NeRRF: 3D Reconstruction and View Synthesis for Transparent and Specular Objects with Neural Refractive-Reflective Fieldscited by 16
  38. 2023CorresNeRF: Image Correspondence Priors for Neural Radiance Fieldscited by 15

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 4; 2026: 6.

11 papers.

  1. 2024Binocular-Guided 3D Gaussian Splatting with View Consistency for Sparse View Synthesiscited by 41
  2. 2025WarpRF: Multi-View Consistency for Training-Free Uncertainty Quantification and Applications in Radiance Fieldscited by 3
  3. 2026OceanSplat: Object-aware Gaussian Splatting with Trinocular View Consistency for Underwater Scene Reconstructioncited by 2
  4. 2026Ego-1K -- A Large-Scale Multiview Video Dataset for Egocentric Visioncited by 2
  5. 2026FreeOcc: Training-Free Embodied Open-Vocabulary Occupancy Predictioncited by 2
  6. 2026In Depth We Trust: Reliable Monocular Depth Supervision for Gaussian Splattingcited by 1
  7. 2026Marigold V2: Revisiting Diffusion Transformers for Monocular Depth Estimationcited by 1
  8. 2025Generating Synthetic Stereo Datasets using 3D Gaussian Splatting and Expert Knowledge Transfercited by 0
  9. 2025Point Cloud Segmentation of Agricultural Vehicles using 3D Gaussian Splattingcited by 0
  10. 2025MatchAttention: Embedding Explicit Matching Constraints into Attention for Efficient Stereo Matchingcited by 0
  11. 2026NeVStereo: A NeRF-Driven NVS-Stereo Architecture for High-Fidelity 3D Taskscited by 0