mlsanityData to 7 Oct 2026

An Adaptive Framework for Learning Unsupervised Depth Completion

arXiv 2106.03010 ↗First posted 6 Jun 2021cs.CV cs.LG cs.ROcited by 30

Not on the map. Why?

Abstract (arXiv)

We present a method to infer a dense depth map from a color image and associated sparse depth measurements. Our main contribution lies in the design of an annealing process for determining co-visibility (occlusions, disocclusions) and the degree of regularization to impose on the model. We show that regularization and co-visibility are related via the fitness (residual) of model to data and both can be unified into a single framework to improve the learning process. Our method is an adaptive weighting scheme that guides optimization by measuring the residual at each pixel location over each training step for (i) estimating a soft visibility mask and (ii) determining the amount of regularization. We demonstrate the effectiveness our method by applying it to several recent unsupervised depth completion methods and improving their performance on public benchmark datasets, without incurring additional trainable parameters or increase in inference time. Code available at: https://github.com/alexklwong/adaframe-depth-completion.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Builds on 31 of its 49 references that are in the map. First posted on arXiv 6 Jun 2021.

Builds on

31 of 49 references.

References · 31

  1. 2017Unsupervised Learning of Depth and Ego-Motion from Videocited by 778
  2. 2017Sparsity Invariant CNNscited by 325
  3. 2018Unsupervised Learning of Depth and Ego-Motion from Monocular Video Using 3D Geometric Constraintscited by 179
  4. 2018Self-supervised Sparse-to-Dense: Self-supervised Depth Completion from LiDAR and Monocular Cameracited by 140
  5. 2018DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene from Sparse LiDAR Data and Single Color Imagecited by 129
  6. 2019DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatchcited by 114
  7. 2018Deep Depth Completion of a Single RGB-D Imagecited by 110
  8. 2019CSPN++: Learning Context and Resource Aware Convolutional Spatial Propagation Networks for Depth Completioncited by 110
  9. 2019Depth Completion from Sparse LiDAR Data with Depth-Normal Constraintscited by 108
  10. 2018In Defense of Classical Image Processing: Fast Depth Completion on the CPUcited by 96
  11. 2019Unsupervised Depth Completion from Visual Inertial Odometrycited by 88
  12. 2019Sparse and noisy LiDAR completion with RGB guidance and uncertaintycited by 79
19 more
  1. 2018Sparse and Dense Data with CNNs: Depth Completion and Semantic Segmentationcited by 76
  2. 2020Learning Joint 2D-3D Representations for Depth Completioncited by 71
  3. 2019Dense Depth Posterior (DDP) from Single Image and Sparse Rangecited by 57
  4. 2019Bilateral Cyclic Constraint and Adaptive Regularization for Unsupervised Monocular Depth Predictioncited by 44
  5. 2020Targeted Adversarial Perturbations for Monocular Depth Predictioncited by 40
  6. 2019DFuseNet: Deep Fusion of RGB and Sparse Depth Information for Image Guided Dense Depth Completioncited by 38
  7. 2018Propagating Confidences through CNNs for Sparse Data Regressioncited by 36
  8. 2018Geo-Supervised Visual Depth Predictioncited by 33
  9. 2018HMS-Net: Hierarchical Multi-scale Sparsity-invariant Network for Sparse Depth Completioncited by 32
  10. 2018Deep Convolutional Compressed Sensing for LiDAR Depth Completioncited by 27
  11. 2018Every Pixel Counts: Unsupervised Geometry Learning with Holistic 3D Motion Understandingcited by 24
  12. 2019Depth Completion via Deep Basis Fittingcited by 23
  13. 2020CodeVIO: Visual-Inertial Odometry with Learned Optimizable Dense Depthcited by 23
  14. 2021Bayesian Deep Basis Fitting for Depth Completion with Uncertaintycited by 23
  15. 2016Improved Stereo Matching with Constant Highway Networks and Reflective Confidence Learningcited by 18
  16. 2020Deep Depth Estimation from Visual-Inertial SLAMcited by 13
  17. 2016Semantically Guided Depth Upsamplingcited by 11
  18. 2016Adaptive Regularization in Convex Composite Optimization for Variational Imaging Problemscited by 4
  19. 2017Adaptive Regularization of Some Inverse Problems in Image Analysiscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 9; 2023: 6; 2024: 5; 2025: 5; 2026: 2.

30 papers.

Sort:

  1. 2021Unsupervised Depth Completion with Calibrated Backprojection Layerscited by 61
  2. 2024WorDepth: Variational Language Prior for Monocular Depth Estimationcited by 34
  3. 2022Deep Depth Completion from Extremely Sparse Data: A Surveycited by 33
  4. 2022Not Just Streaks: Towards Ground Truth for Single Image Derainingcited by 31
  5. 2022DesNet: Decomposed Scale-Consistent Network for Unsupervised Depth Completioncited by 27
  6. 2024Test-Time Adaptation for Depth Completioncited by 26
  7. 2022Monitored Distillation for Positive Congruent Depth Completioncited by 24
  8. 2022Multi-Modal Masked Pre-Training for Monocular Panoramic Depth Completioncited by 23
  9. 2024RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptionscited by 20
  10. 2021Stereoscopic Universal Perturbations across Different Architectures and Datasetscited by 17
  11. 2021Robust Depth Completion with Uncertainty-Driven Loss Functionscited by 16
  12. 2023AugUndo: Scaling Up Augmentations for Monocular Depth Completion and Estimationcited by 15
18 more
  1. 2023Learnable Differencing Center for Nighttime Depth Perceptioncited by 14
  2. 2023Heteroscedastic Uncertainty Estimation Framework for Unsupervised Registrationcited by 14
  3. 2025ProtoDepth: Unsupervised Continual Depth Completion with Prototypescited by 11
  4. 2023Adaptive Correspondence Scoring for Unsupervised Medical Image Registrationcited by 9
  5. 2025Extending Foundational Monocular Depth Estimators to Fisheye Cameras with Calibration Tokenscited by 9
  6. 2024All-day Depth Completioncited by 8
  7. 2024UnCLe: Benchmarking Unsupervised Continual Learning for Depth Completioncited by 5
  8. 2025Radar-Guided Polynomial Fitting for Metric Depth Estimationcited by 5
  9. 2025ETA: Energy-based Test-time Adaptation for Depth Completioncited by 4
  10. 2022Segmentation-guided Domain Adaptation for Efficient Depth Completioncited by 2
  11. 2022FCDSN-DC: An Accurate and Lightweight Convolutional Neural Network for Stereo Estimation with Depth Completioncited by 1
  12. 2025StarryGazer: Leveraging Monocular Depth Estimation Models for Domain-Agnostic Single Depth Image Completioncited by 1
  13. 2026VisTa3D: A Dataset and Benchmark for Thin Object Reconstruction from Vision, Tactile, and 3D Point Cloudscited by 1
  14. 2022On the Viability of Monocular Depth Pre-training for Semantic Segmentationcited by 0
  15. 2022MetaComp: Learning to Adapt for Online Depth Completioncited by 0
  16. 2023DEUX: Active Exploration for Learning Unsupervised Depth Perceptioncited by 0
  17. 2023Self-Supervised Depth Completion Guided by 3D Perception and Geometry Consistencycited by 0
  18. 2026From Perspective to Fisheye Depth Estimation and Open-Vocabulary Segmentationcited by 0