mlsanityData to 7 Oct 2026

Input-level Inductive Biases for 3D Reconstruction

arXiv 2112.03243 ↗First posted 6 Dec 2021cs.CVcited by 31

Not on the map. Why?

Abstract (arXiv)

Much of the recent progress in 3D vision has been driven by the development of specialized architectures that incorporate geometrical inductive biases. In this paper we tackle 3D reconstruction using a domain agnostic architecture and study how instead to inject the same type of inductive biases directly as extra inputs to the model. This approach makes it possible to apply existing general models, such as Perceivers, on this rich domain, without the need for architectural changes, while simultaneously maintaining data efficiency of bespoke models. In particular we study how to encode cameras, projective ray incidence and epipolar geometry as model inputs, and demonstrate competitive multi-view depth estimation performance on multiple benchmarks.

Key facts. Cited by 31 papers in the map (data through 7 Oct 2026). Builds on 52 of its 70 references that are in the map. First posted on arXiv 6 Dec 2021.

Builds on

52 of 70 references.

References · 52

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  5. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  6. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  7. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  8. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  9. 2014How transferable are features in deep neural networks?cited by 1,216
  10. 2021Perceiver: General Perception with Iterative Attentioncited by 1,089
  11. 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
  12. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,054
38 more
  1. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 974
  2. 2017PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volumecited by 966
  3. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  4. 2021Perceiver IO: A General Architecture for Structured Inputs & Outputscited by 616
  5. 2018Pyramid Stereo Matching Networkcited by 486
  6. 2019Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matchingcited by 414
  7. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 359
  8. 2014CNN Features off-the-shelf: an Astounding Baseline for Recognitioncited by 316
  9. 2018DeepMVS: Learning Multi-view Stereopsiscited by 277
  10. 2019What Do Single-view 3D Reconstruction Networks Learn?cited by 259
  11. 2019GA-Net: Guided Aggregation Net for End-to-end Stereo Matchingcited by 242
  12. 2021NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Videocited by 241
  13. 2019Group-wise Correlation Stereo Networkcited by 235
  14. 2020Deep Global Registrationcited by 228
  15. 2020Atlas: End-to-End 3D Scene Reconstruction from Posed Imagescited by 206
  16. 2020AANet: Adaptive Aggregation Network for Efficient Stereo Matchingcited by 202
  17. 2016DeMoN: Depth and Motion Network for Learning Monocular Stereocited by 200
  18. 2020Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective with Transformerscited by 190
  19. 2021COTR: Correspondence Transformer for Matching Across Imagescited by 182
  20. 2017Learning to Find Good Correspondencescited by 176
  21. 2020CrossTransformers: spatially-aware few-shot transfercited by 174
  22. 2018Neighbourhood Consensus Networkscited by 173
  23. 2020Hierarchical Neural Architecture Search for Deep Stereo Matchingcited by 166
  24. 2019Cost Volume Pyramid Based Depth Inference for Multi-View Stereocited by 154
  25. 2016Deep Image Homography Estimationcited by 135
  26. 2017SurfaceNet: An End-to-end 3D Neural Network for Multiview Stereopsiscited by 128
  27. 2019Hierarchical Deep Stereo Matching on High-resolution Imagescited by 126
  28. 2020Epipolar Transformerscited by 117
  29. 2019DPSNet: End-to-end Deep Plane Sweep Stereocited by 115
  30. 2021TransformerFusion: Monocular RGB Scene Reconstruction using Transformerscited by 107
  31. 2020Efficient Neighbourhood Consensus Networks via Submanifold Sparse Convolutionscited by 92
  32. 2018DeepTAM: Deep Tracking and Mappingcited by 87
  33. 2020Guiding Monocular Depth Estimation Using Depth-Attention Volumecited by 82
  34. 2021Learning Signed Distance Field for Multi-view Surface Reconstructioncited by 76
  35. 2019CAM-Convs: Camera-Aware Multi-Scale Convolutions for Single-View Depthcited by 75
  36. 2018DGC-Net: Dense Geometric Correspondence Networkcited by 67
  37. 2019Hyperpixel Flow: Semantic Correspondence with Multi-layer Neural Featurescited by 56
  38. 2019Normal Assisted Stereo Depth Estimationcited by 49

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 7; 2023: 8; 2024: 6; 2025: 7; 2026: 3.

31 papers.

Sort:

By area:

  1. 2024Grounding Image Matching in 3D with MASt3Rcited by 960
  2. 2022Unifying Flow, Stereo and Depth Estimationcited by 270
  3. 2023Learning to Render Novel Views from Wide-Baseline Stereo Pairscited by 103
  4. 2023Towards Zero-Shot Scale-Aware Monocular Depth Estimationcited by 100
  5. 2023Long-Term Photometric Consistent Novel View Synthesis with Diffusion Modelscited by 76
  6. 2025Dens3R: A Foundation Model for 3D Geometry Predictioncited by 44
  7. 2022The 8-Point Algorithm as an Inductive Bias for Relative Pose Prediction by ViTscited by 38
  8. 20233D Hand Pose Estimation in Everyday Egocentric Imagescited by 33
  9. 2022LaRa: Latents and Rays for Multi-Camera Bird's-Eye-View Semantic Segmentationcited by 29
  10. 2022Attention-based Neural Cellular Automatacited by 26
  11. 2024Mitigating Covariate Shift in Imitation Learning for Autonomous Vehicles Using Latent Space Generative World Modelscited by 22
  12. 2023Viewpoint Equivariance for Multi-View 3D Object Detectioncited by 18
19 more
  1. 2022Depth Field Networks for Generalizable Multi-view Scene Representationcited by 17
  2. 2023Pixel-Aligned Recurrent Queries for Multi-View 3D Object Detectioncited by 14
  3. 2025Do You Know Where Your Camera Is? View-Invariant Policy Learning with Camera Conditioningcited by 14
  4. 2025Zero-Shot Novel View and Depth Synthesis with Multi-View Geometric Diffusioncited by 12
  5. 2026RayRoPE: Projective Ray Positional Encoding for Multi-view Attentioncited by 10
  6. 2022A Light Touch Approach to Teaching Transformers Multi-view Geometrycited by 6
  7. 2024GRIN: Zero-Shot Metric Depth with Pixel-Level Diffusioncited by 6
  8. 2024SE(3) Equivariant Ray Embeddings for Implicit Multi-View Depth Estimationcited by 6
  9. 2023DeLiRa: Self-Supervised Depth, Light, and Radiance Fieldscited by 5
  10. 2026Boxer: Robust Lifting of Open-World 2D Bounding Boxes to 3Dcited by 5
  11. 2023Mitigating Perspective Distortion-induced Shape Ambiguity in Image Cropscited by 4
  12. 2025PanMatch: Unleashing the Potential of Large Vision Models for Unified Matching Modelscited by 2
  13. 2025Towards Depth Foundation Model: Recent Trends in Vision-Based Depth Estimationcited by 2
  14. 2024SCENES: Subpixel Correspondence Estimation With Epipolar Supervisioncited by 1
  15. 2024Exploiting Motion Prior for Accurate Pose Estimation of Dashboard Camerascited by 1
  16. 2025A Study of Finetuning Video Transformers for Multi-view Geometry Taskscited by 1
  17. 2022Neural Camera Modelscited by 0
  18. 2025Embodied Spatial Intelligence: from Implicit Scene Modeling to Spatial Reasoningcited by 0
  19. 2026GIBLy: Improving 3D Semantic Segmentation through an Architecture-Agnostic Lightweight Geometric Inductive Bias Layercited by 0