mlsanityData to 7 Oct 2026

IronDepth: Iterative Refinement of Single-View Depth using Surface Normal and its Uncertainty

arXiv 2210.03676 ↗First posted 7 Oct 2022cs.CVcited by 30

Not on the map. Why?

Abstract (arXiv)

Single image surface normal estimation and depth estimation are closely related problems as the former can be calculated from the latter. However, the surface normals computed from the output of depth estimation methods are significantly less accurate than the surface normals directly estimated by networks. To reduce such discrepancy, we introduce a novel framework that uses surface normal and its uncertainty to recurrently refine the predicted depth-map. The depth of each pixel can be propagated to a query pixel, using the predicted surface normal as guidance. We thus formulate depth refinement as a classification of choosing the neighboring pixel to propagate from. Then, by propagating to sub-pixel points, we upsample the refined, low-resolution output. The proposed method shows state-of-the-art performance on NYUv2 and iBims-1 - both in terms of depth and normal. Our refinement module can also be attached to the existing depth estimation methods to improve their accuracy. We also show that our framework, only trained for depth estimation, can also be used for depth completion. The code is available at https://github.com/baegwangbin/IronDepth.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Builds on 30 of its 37 references that are in the map. First posted on arXiv 7 Oct 2022.

Builds on

30 of 37 references.

References · 30

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2020End-to-End Object Detection with Transformerscited by 7,384
  4. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  5. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  6. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  7. 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,303
  8. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,264
  9. 2018Deep Ordinal Regression Network for Monocular Depth Estimationcited by 721
  10. 2020AdaBins: Depth Estimation using Adaptive Binscited by 596
  11. 2014Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecturecited by 483
  12. 2016Deeper Depth Prediction with Fully Convolutional Residual Networkscited by 402
18 more
  1. 2019From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimationcited by 379
  2. 2019Enforcing geometric constraints of virtual normal for depth predictioncited by 233
  3. 2017Super-Convergence: Very Fast Training of Neural Networks Using Large Learning Ratescited by 196
  4. 2018Evaluation of CNN-based Single-Image Depth Estimation Methodscited by 137
  5. 2021Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimationcited by 116
  6. 2017Semi-Supervised Deep Learning for Monocular Depth Map Predictioncited by 109
  7. 2014Deep Convolutional Neural Fields for Depth Estimation from a Single Imagecited by 106
  8. 2016Estimating Depth from Monocular Images as Classification Using Deep Fully Convolutional Residual Networkscited by 91
  9. 2020Guiding Monocular Depth Estimation Using Depth-Attention Volumecited by 82
  10. 2014Designing Deep Networks for Surface Normal Estimationcited by 59
  11. 2019SharpNet: Fast and Accurate Recovery of Occluding Contours in Monocular Depth Estimationcited by 59
  12. 2020One Shot 3D Photographycited by 45
  13. 2021Adaptive Surface Normal Constraint for Depth Estimationcited by 43
  14. 2019Spherical Regression: Learning Viewpoints, Surface Normals and 3D Rotations on n-Spherescited by 37
  15. 2020GeoNet++: Iterative Geometric Neural Network with Edge-Aware Refinement for Joint Depth and Surface Normal Estimationcited by 36
  16. 2016Marr Revisited: 2D-3D Alignment via Surface Normal Predictioncited by 34
  17. 2020Surface Normal Estimation of Tilted Images via Spatial Rectifiercited by 34
  18. 2019FrameNet: Learning Local Canonical Frames of 3D Surfaces from a Single RGB Imagecited by 26

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 11; 2024: 13; 2025: 2; 2026: 4.

30 papers.

Sort:

By area:

  1. 2024Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimationcited by 389
  2. 2024GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Imagecited by 283
  3. 2023Text2Room: Extracting Textured 3D Meshes from 2D Text-to-Image Modelscited by 245
  4. 2024Rethinking Inductive Biases for Surface Normal Estimationcited by 122
  5. 2023iDisc: Internal Discretization for Monocular Depth Estimationcited by 71
  6. 2023DiffCAD: Weakly-Supervised Probabilistic CAD Model Retrieval and Alignment from an RGB Imagecited by 48
  7. 2024Invisible Stitch: Generating Smooth 3D Scenes with Depth Inpaintingcited by 23
  8. 2024OGNI-DC: Robust Depth Completion with Optimization-Guided Neural Iterationscited by 23
  9. 2023UncLe-SLAM: Uncertainty Learning for Dense Neural SLAMcited by 19
  10. 2023Learning a Depth Covariance Functioncited by 16
  11. 2023Measured Albedo in the Wild: Filling the Gap in Intrinsics Evaluationcited by 16
  12. 2023Single Image Depth Prediction Made Better: A Multivariate Gaussian Takecited by 14
18 more
  1. 2024Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledgecited by 14
  2. 2024Leveraging Near-Field Lighting for Monocular Depth Estimation from Endoscopy Videoscited by 14
  3. 2024GenRC: Generative 3D Room Completion from Sparse Image Collectionscited by 9
  4. 2023Meta-Optimization for Higher Model Generalizability in Single-Image Depth Predictioncited by 3
  5. 2023U2RLE: Uncertainty-Guided 2-Stage Room Layout Estimationcited by 2
  6. 2024Prim2Room: Layout-Controllable Room Mesh Generation from Primitivescited by 2
  7. 2026LiftFormer: Lifting and Frame Theory Based Monocular Depth Estimation Using Depth and Edge Oriented Subspace Representationcited by 2
  8. 2023InSpaceType: Reconsider Space Type in Indoor Monocular Depth Estimationcited by 1
  9. 2024MultiDepth: Multi-Sample Priors for Refining Monocular Metric Depth Estimations in Indoor Scenescited by 1
  10. 2025HiGS: Hierarchical Generative Scene Framework for Multi-Step Associative Semantic Spatial Compositioncited by 1
  11. 2023VioLA: Aligning Videos to 2D LiDAR Scanscited by 0
  12. 2024MEDeA: Multi-view Efficient Depth Adjustmentcited by 0
  13. 2024InSpaceType: Dataset and Benchmark for Reconsidering Cross-Space Type Performance in Indoor Monocular Depthcited by 0
  14. 2024A Practical Approach to Underwater Depth and Surface Normals Estimationcited by 0
  15. 2025M3Depth: Wavelet-Enhanced Depth Estimation on Mars via Mutual Boosting of Dual-Modal Datacited by 0
  16. 2026Sparse-View Surface Reconstruction using Gaussian Splatting through High-Confidence Depth Propagation with Normal Priorscited by 0
  17. 2026Interpretable Human-Label-Free Deep Learning for Real-Bogus Classification with Uncertainty Quantificationcited by 0
  18. 2026TransNormal-2: Geometry-Grounded Rectified Flow with Edge-Aware Decoding for Precise Normal Estimationcited by 0