mlsanityData to 7 Oct 2026

SceneRF: Self-Supervised Monocular 3D Scene Reconstruction with Radiance Fields

arXiv 2212.02501 ↗First posted 5 Dec 2022cs.CV cs.AI cs.GRcited by 49

Not on the map. Why?

Abstract (arXiv)

3D reconstruction from a single 2D image was extensively covered in the literature but relies on depth supervision at training time, which limits its applicability. To relax the dependence to depth we propose SceneRF, a self-supervised monocular scene reconstruction method using only posed image sequences for training. Fueled by the recent progress in neural radiance fields (NeRF) we optimize a radiance field though with explicit depth optimization and a novel probabilistic sampling strategy to efficiently handle large scenes. At inference, a single input image suffices to hallucinate novel depth views which are fused together to obtain 3D scene reconstruction. Thorough experiments demonstrate that we outperform all baselines for novel depth views synthesis and scene reconstruction, on indoor BundleFusion and outdoor SemanticKITTI. Code is available at https://astra-vision.github.io/SceneRF .

Key facts. Cited by 49 papers in the map (data through 7 Oct 2026). Builds on 60 of its 101 references that are in the map. First posted on arXiv 5 Dec 2022.

Builds on

60 of 101 references.

References · 60

  1. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  2. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  3. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  4. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  5. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  6. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,013
  7. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,574
  8. 2016A Point Set Generation Network for 3D Object Reconstruction from a Single Imagecited by 972
  9. 2019Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representationscited by 864
  10. 2016Unsupervised Monocular Depth Estimation with Left-Right Consistencycited by 819
  11. 2020Convolutional Occupancy Networkscited by 760
  12. 2021Depth-supervised NeRF: Fewer Views and Faster Training for Freecited by 732
38 more
  1. 2021Resolution-robust Large Mask Inpainting with Fourier Convolutionscited by 730
  2. 2019Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervisioncited by 699
  3. 2019Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelinescited by 672
  4. 2018Pixel2Mesh: Generating 3D Mesh Models from Single RGB Imagescited by 649
  5. 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 646
  6. 2020AdaBins: Depth Estimation using Adaptive Binscited by 596
  7. 2021Neural Fields in Visual Computing and Beyondcited by 539
  8. 2019PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flowscited by 487
  9. 2019SynSin: End-to-end View Synthesis from a Single Imagecited by 364
  10. 2021MonoScene: Monocular 3D Semantic Scene Completioncited by 357
  11. 20163DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructionscited by 333
  12. 2019DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstructioncited by 332
  13. 2021Light Field Networks: Neural Scene Representations with Single-Evaluation Renderingcited by 330
  14. 2021Dense Depth Priors for Neural Radiance Fields from Sparse Input Viewscited by 301
  15. 2021Urban Radiance Fieldscited by 263
  16. 2021DONeRF: Towards Real-Time Rendering of Compact Neural Radiance Fields using Depth Oracle Networkscited by 232
  17. 2019BSP-Net: Generating Compact Meshes via Binary Space Partitioningcited by 205
  18. 2020LMSCNet: Lightweight Multiscale 3D Semantic Completioncited by 187
  19. 2021CodeNeRF: Disentangled Neural Radiance Fields for Object Categoriescited by 183
  20. 2020Total3DUnderstanding: Joint Layout, Object Pose and Mesh Reconstruction for Indoor Scenes from a Single Imagecited by 170
  21. 2020Perceiving 3D Human-Object Spatial Arrangements from a Single Image in the Wildcited by 152
  22. 2021MINE: Towards Continuous Depth MPI with NeRF for Novel View Synthesiscited by 133
  23. 20193D Ken Burns Effect from a Single Imagecited by 128
  24. 2020Anisotropic Convolutional Networks for 3D Semantic Scene Completioncited by 126
  25. 20203D Sketch-aware Semantic Scene Completion via Semi-supervised Structure Priorcited by 124
  26. 2022EfficientNeRF: Efficient Neural Radiance Fieldscited by 110
  27. 2019HorizonNet: Learning Room Layout with 1D Representation and Pano Stretch Data Augmentationcited by 107
  28. 2022Vision Transformer for NeRF-Based View Synthesis from a Single Input Imagecited by 104
  29. 2018LayoutNet: Reconstructing the 3D Room Layout from a Single RGB Imagecited by 94
  30. 2021Holistic 3D Scene Understanding from a Single Image with Implicit Representationcited by 94
  31. 2021ShaRF: Shape-conditioned Radiance Fields from a Single Viewcited by 92
  32. 2020Pix2Vox++: Multi-scale Context-aware 3D Object Reconstruction from Single and Multiple Imagescited by 79
  33. 2016IM2CADcited by 78
  34. 20213D Semantic Scene Completion: a Surveycited by 74
  35. 2021Panoptic 3D Scene Reconstruction From a Single RGB Imagecited by 70
  36. 2018Holistic 3D Scene Parsing and Reconstruction from a Single RGB Imagecited by 69
  37. 2022AdaNeRF: Adaptive Sampling for Real-time Rendering of Neural Radiance Fieldscited by 68
  38. 2022AutoRF: Learning 3D Object Radiance Fields from Single View Observationscited by 59

50 most cited of 60 All 60 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 14; 2024: 15; 2025: 15; 2026: 5.

49 papers.

Sort:

By area:

  1. 2023OccWorld: Learning a 3D Occupancy World Model for Autonomous Drivingcited by 230
  2. 2023Scene as Occupancycited by 208
  3. 2023SelfOcc: Self-Supervised Vision-Based 3D Occupancy Predictioncited by 133
  4. 2023OccDepth: A Depth-Aware Method for 3D Semantic Scene Completioncited by 85
  5. 2024GaussianFormer-2: Probabilistic Gaussian Superposition for Efficient 3D Occupancy Predictioncited by 83
  6. 2024TIP-Editor: An Accurate 3D Editor Following Both Text-Prompts And Image-Promptscited by 76
  7. 2023Behind the Scenes: Density Fields for Single View Reconstructioncited by 63
  8. 2024GaussianOcc: Fully Self-supervised and Efficient 3D Occupancy Estimation with Gaussian Splattingcited by 50
  9. 2024GaussTR: Foundation Model-Aligned Gaussian Transformer for Self-Supervised 3D Spatial Understandingcited by 46
  10. 2023OccNeRF: Advancing 3D Occupancy Prediction in LiDAR-Free Environmentscited by 39
  11. 2023MonoNeRD: NeRF-like Representations for Monocular 3D Object Detectioncited by 33
  12. 2023MUVO: A Multimodal Generative World Model for Autonomous Driving with Geometric Representationscited by 33
37 more
  1. 2023S4C: Self-Supervised Semantic Scene Completion with Neural Fieldscited by 27
  2. 2023Grid-Centric Traffic Scenario Perception for Autonomous Driving: A Comprehensive Reviewcited by 25
  3. 2024Let Occ Flow: Self-Supervised 3D Occupancy Flow Predictioncited by 23
  4. 2025DVGT: Driving Visual Geometry Transformercited by 20
  5. 2025QuadricFormer: Scene as Superquadrics for 3D Semantic Occupancy Predictioncited by 19
  6. 2024Neural Radiance Field-based Visual Rendering: A Comprehensive Reviewcited by 12
  7. 2024HybridOcc: NeRF Enhanced Transformer-based Multi-Camera 3D Occupancy Predictioncited by 12
  8. 2024SurroundSDF: Implicit 3D Scene Understanding Based on Signed Distance Fieldcited by 10
  9. 2024OccScene: Semantic Occupancy-based Cross-task Mutual Learning for 3D Scene Generationcited by 10
  10. 2024Semantically-aware Neural Radiance Fields for Visual Scene Understanding: A Comprehensive Reviewcited by 9
  11. 2025AGO: Adaptive Grounding for Open World 3D Occupancy Predictioncited by 9
  12. 2026OccAny: Generalized Unconstrained Urban 3D Occupancycited by 6
  13. 2024PyTorchGeoNodes: Enabling Differentiable Shape Programs for 3D Shape Reconstructioncited by 5
  14. 2024Semi-supervised 3D Semantic Scene Completion with 2D Vision Foundation Model Guidancecited by 5
  15. 2025SceneCompleter: Dense 3D Scene Completion for Generative Novel View Synthesiscited by 5
  16. 2023A Simple Framework for 3D Occupancy Estimation in Autonomous Drivingcited by 4
  17. 2023Federated Multi-View Synthesizing for Metaversecited by 4
  18. 2024Gated Fields: Learning Scene Reconstruction from Gated Videoscited by 3
  19. 2025H3O: Hyper-Efficient 3D Occupancy Prediction with Heterogeneous Supervisioncited by 3
  20. 2025ODG: Occupancy Prediction Using Dual Gaussianscited by 3
  21. 2023Novel View Synthesis with View-Dependent Effects from a Single Imagecited by 2
  22. 2025Diffusion-Based Generative Models for 3D Occupancy Prediction in Autonomous Drivingcited by 2
  23. 2026To View Transform or Not to View Transform: NeRF-based Pre-training Perspectivecited by 2
  24. 2025EA-3DGS: Efficient and Adaptive 3D Gaussians with Highly Enhanced Quality for outdoor scenescited by 1
  25. 2025BePo: Dual Representation for 3D Occupancy Predictioncited by 1
  26. 2025Dream-to-Recon: Monocular 3D Reconstruction with Diffusion-Depth Distillation from Single Imagescited by 1
  27. 2025GaussEdit: Adaptive 3D Scene Editing with Text and Image Promptscited by 1
  28. 2025ImagineNav++: Prompting Vision-Language Models as Embodied Navigator through Scene Imaginationcited by 1
  29. 2023Survey on Controlable Image Synthesis with Deep Learningcited by 0
  30. 2024Neural Radiance Fields with Torch Unitscited by 0
  31. 2024VoxDepth: Rectification of Depth Images on Edge Devicescited by 0
  32. 2025Learning A Zero-shot Occupancy Network from Vision Foundation Models via Self-supervised Adaptationcited by 0
  33. 2025Image-Plane Geometric Decoding for View-Invariant Indoor Scene Reconstructioncited by 0
  34. 2025Asset-Driven Sematic Reconstruction of Dynamic Scene with Multi-Human-Object Interactionscited by 0
  35. 2026Rebenchmarking Unsupervised Monocular 3D Occupancy Predictioncited by 0
  36. 2026SelfOccFlow: Towards end-to-end self-supervised 3D Occupancy Flow predictioncited by 0
  37. 2026DecoRec: Decomposed 3D Scene Reconstruction from Single-View Images via Object-Level Diffusioncited by 0