mlsanityData to 7 Oct 2026

DualRefine: Self-Supervised Depth and Pose Estimation Through Iterative Epipolar Sampling and Refinement Toward Equilibrium

arXiv 2304.03560 ↗First posted 7 Apr 2023cs.CVcited by 16

Not on the map. Why?

Abstract (arXiv)

Self-supervised multi-frame depth estimation achieves high accuracy by computing matching costs of pixel correspondences between adjacent frames, injecting geometric information into the network. These pixel-correspondence candidates are computed based on the relative pose estimates between the frames. Accurate pose predictions are essential for precise matching cost computation as they influence the epipolar geometry. Furthermore, improved depth estimates can, in turn, be used to align pose estimates. Inspired by traditional structure-from-motion (SfM) principles, we propose the DualRefine model, which tightly couples depth and pose estimation through a feedback loop. Our novel update pipeline uses a deep equilibrium model framework to iteratively refine depth estimates and a hidden state of feature maps by computing local matching costs based on epipolar geometry. Importantly, we used the refined depth estimates and feature maps to compute pose updates at each step. This update in the pose estimates slowly alters the epipolar geometry during the refinement process. Experimental results on the KITTI dataset demonstrate competitive depth prediction and odometry prediction performance surpassing published self-supervised baselines.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 89 of its 112 references that are in the map. First posted on arXiv 7 Apr 2023.

Builds on

89 of 112 references.

References · 89

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  4. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  5. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  6. 2021Image Super-Resolution via Iterative Refinementcited by 1,506
  7. 2015Spatial Transformer Networkscited by 1,423
  8. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,417
  9. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,264
  10. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  11. 2016Unsupervised Monocular Depth Estimation with Left-Right Consistencycited by 819
  12. 2016ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Camerascited by 801
38 more
  1. 2018MVSNet: Depth Inference for Unstructured Multi-view Stereocited by 780
  2. 2017Unsupervised Learning of Depth and Ego-Motion from Videocited by 778
  3. 2018MegaDepth: Learning Single-View Depth Prediction from Internet Photoscited by 731
  4. 2018Deep Ordinal Regression Network for Monocular Depth Estimationcited by 721
  5. 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 673
  6. 2018Taskonomy: Disentangling Task Transfer Learningcited by 631
  7. 2019Deep Equilibrium Modelscited by 603
  8. 2014Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecturecited by 483
  9. 2016Direct Sparse Odometrycited by 479
  10. 2016RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentationcited by 463
  11. 20193D Packing for Self-Supervised Monocular Depth Estimationcited by 437
  12. 2016Deeper Depth Prediction with Fully Convolutional Residual Networkscited by 402
  13. 2019From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimationcited by 379
  14. 2016Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescuecited by 368
  15. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 359
  16. 2018GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Posecited by 341
  17. 2017Sparsity Invariant CNNscited by 325
  18. 2018DeepMVS: Learning Multi-view Stereopsiscited by 277
  19. 2019Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Videocited by 244
  20. 2021Learning to Estimate Hidden Motions with Global Motion Aggregationcited by 223
  21. 2018Competitive Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentationcited by 222
  22. 2020Consistent Video Depth Estimationcited by 220
  23. 2019CANet: Class-Agnostic Segmentation Networks with Iterative Refinement and Attentive Few-Shot Learningcited by 217
  24. 2020D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometrycited by 206
  25. 2016DeMoN: Depth and Motion Network for Learning Monocular Stereocited by 200
  26. 2021Back to the Feature: Learning Robust Camera Localization from Pixels to Posecited by 199
  27. 2021NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereocited by 191
  28. 2021The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depthcited by 190
  29. 2018DeepV2D: Video to Depth with Differentiable Structure from Motioncited by 185
  30. 2016Single-Image Depth Perception in the Wildcited by 179
  31. 2018Unsupervised Learning of Depth and Ego-Motion from Monocular Video Using 3D Geometric Constraintscited by 179
  32. 2019Depth from Videos in the Wild: Unsupervised Monocular Depth Learning from Unknown Camerascited by 170
  33. 2018Unsupervised Learning of Monocular Depth Estimation and Visual Odometry with Deep Feature Reconstructioncited by 159
  34. 2018Depth Prediction Without the Sensors: Leveraging Structure for Unsupervised Learning from Monocular Videoscited by 155
  35. 2017Learning Depth from Monocular Videos using Direct Methodscited by 154
  36. 2018DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistencycited by 148
  37. 2020Self-Supervised Monocular Depth Estimation: Solving the Dynamic Object Problem by Semantic Guidancecited by 148
  38. 2020Feature-metric Loss for Self-supervised Learning of Depth and Egomotioncited by 140

50 most cited of 89 All 89 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 8; 2025: 5; 2026: 2.

16 papers.

Sort:

By area:

  1. 2024Cross-View Completion Models are Zero-shot Correspondence Estimatorscited by 30
  2. 2025Jasmine: Harnessing Diffusion Prior for Self-supervised Depth Estimationcited by 20
  3. 2025Synthetic-to-Real Self-supervised Robust Depth Estimation via Learning with Motion and Structure Priorscited by 20
  4. 2024Mono-ViFI: A Unified Learning Framework for Self-supervised Single- and Multi-frame Monocular Depth Estimationcited by 9
  5. 2024Embodiment: Self-Supervised Depth Estimation Based on Camera Modelscited by 9
  6. 2024ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusioncited by 6
  7. 2025DINO-VO: A Feature-based Visual Odometry Leveraging a Visual Foundation Modelcited by 5
  8. 2023AltNeRF: Learning Robust Neural Radiance Field via Alternating Depth-Pose Optimizationcited by 4
  9. 2024MAL: Motion-Aware Loss with Temporal and Distillation Hints for Self-Supervised Depth Estimationcited by 4
  10. 2025CoProU-VO: Combining Projected Uncertainty for End-to-End Unsupervised Monocular Visual Odometrycited by 1
  11. 2024FA-Depth: Toward Fast and Accurate Self-supervised Monocular Depth Estimationcited by 0
  12. 2024Str-L Pose: Integrating Point and Structured Line for Relative Pose Estimation in Dual-Graphcited by 0
4 more
  1. 2024Self-supervised Monocular Depth and Pose Estimation for Endoscopy with Latent Priorscited by 0
  2. 2025Discriminately Treating Motion Components Evolves Joint Depth and Ego-Motion Learningcited by 0
  3. 2026SA4Depth: Consistent Pose-Depth Scale Alignment for Self-Supervised Monocular Depth Estimationcited by 0
  4. 2026Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimationcited by 0