mlsanityData to 14 Oct 2026

Learning Topology from Synthetic Data for Unsupervised Depth Completion

arXiv 2106.02994 ↗First posted 6 Jun 2021cs.CV cs.LG cs.ROcited by 38

Not on the map. Why?

Abstract (arXiv)

We present a method for inferring dense depth maps from images and sparse depth measurements by leveraging synthetic data to learn the association of sparse point clouds with dense natural shapes, and using the image as evidence to validate the predicted depth map. Our learned prior for natural shapes uses only sparse depth as input, not images, so the method is not affected by the covariate shift when attempting to transfer learned models from synthetic data to real ones. This allows us to use abundant synthetic data with ground truth to learn the most difficult component of the reconstruction process, which is topology estimation, and use the image to refine the prediction based on photometric evidence. Our approach uses fewer parameters than previous methods, yet, achieves the state of the art on both indoor and outdoor benchmark datasets. Code available at: https://github.com/alexklwong/learning-topology-synthetic-data.

Key facts. Cited by 38 papers in the map (data through 14 Oct 2026). Builds on 31 of its 48 references that are in the map. First posted on arXiv 6 Jun 2021.

Builds on

31 of 48 references.

References · 31

  1. 2014Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognitioncited by 720
  2. 2017Understanding the Effective Receptive Field in Deep Convolutional Neural Networkscited by 553
  3. 2018Pyramid Stereo Matching Networkcited by 489
  4. 20193D Packing for Self-Supervised Monocular Depth Estimationcited by 439
  5. 2016Virtual Worlds as Proxy for Multi-Object Tracking Analysiscited by 341
  6. 2017Sparsity Invariant CNNscited by 326
  7. 2018Self-supervised Sparse-to-Dense: Self-supervised Depth Completion from LiDAR and Monocular Cameracited by 140
  8. 2018DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene from Sparse LiDAR Data and Single Color Imagecited by 129
  9. 2018Deep Depth Completion of a Single RGB-D Imagecited by 110
  10. 2019CSPN++: Learning Context and Resource Aware Convolutional Spatial Propagation Networks for Depth Completioncited by 110
  11. 2019Depth Completion from Sparse LiDAR Data with Depth-Normal Constraintscited by 108
  12. 2018In Defense of Classical Image Processing: Fast Depth Completion on the CPUcited by 96
19 more
  1. 2019Unsupervised Depth Completion from Visual Inertial Odometrycited by 89
  2. 2019Sparse and noisy LiDAR completion with RGB guidance and uncertaintycited by 79
  3. 2018Sparse and Dense Data with CNNs: Depth Completion and Semantic Segmentationcited by 76
  4. 2020Learning Joint 2D-3D Representations for Depth Completioncited by 71
  5. 2019Dense Depth Posterior (DDP) from Single Image and Sparse Rangecited by 57
  6. 2018AdaDepth: Unsupervised Content Congruent Adaptation for Depth Estimationcited by 54
  7. 2019Bilateral Cyclic Constraint and Adaptive Regularization for Unsupervised Monocular Depth Predictioncited by 44
  8. 2020Targeted Adversarial Perturbations for Monocular Depth Predictioncited by 40
  9. 2019DFuseNet: Deep Fusion of RGB and Sparse Depth Information for Image Guided Dense Depth Completioncited by 38
  10. 2018Propagating Confidences through CNNs for Sparse Data Regressioncited by 36
  11. 2018Geo-Supervised Visual Depth Predictioncited by 33
  12. 2016SceneNet RGB-D: 5M Photorealistic Images of Synthetic Indoor Trajectories with Ground Truthcited by 32
  13. 2018HMS-Net: Hierarchical Multi-scale Sparsity-invariant Network for Sparse Depth Completioncited by 32
  14. 2018Deep Convolutional Compressed Sensing for LiDAR Depth Completioncited by 27
  15. 2019Depth Completion via Deep Basis Fittingcited by 23
  16. 2020CodeVIO: Visual-Inertial Odometry with Learned Optimizable Dense Depthcited by 23
  17. 2021Bayesian Deep Basis Fitting for Depth Completion with Uncertaintycited by 23
  18. 2020Deep Depth Estimation from Visual-Inertial SLAMcited by 13
  19. 2016Semantically Guided Depth Upsamplingcited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 2; 2022: 13; 2023: 7; 2024: 7; 2025: 7; 2026: 2.

38 papers.

Sort:

By area:

  1. 2021Unsupervised Depth Completion with Calibrated Backprojection Layerscited by 61
  2. 2023LRRU: Long-short Range Recurrent Updating Networks for Depth Completioncited by 49
  3. 2024WorDepth: Variational Language Prior for Monocular Depth Estimationcited by 34
  4. 2022Deep Depth Completion from Extremely Sparse Data: A Surveycited by 33
  5. 2022Not Just Streaks: Towards Ground Truth for Single Image Derainingcited by 31
  6. 2022DesNet: Decomposed Scale-Consistent Network for Unsupervised Depth Completioncited by 27
  7. 2022Sparsity Agnostic Depth Completioncited by 27
  8. 2024Test-Time Adaptation for Depth Completioncited by 26
  9. 2022Monitored Distillation for Positive Congruent Depth Completioncited by 24
  10. 2022Multi-Modal Masked Pre-Training for Monocular Panoramic Depth Completioncited by 23
  11. 2022ALTO: Alternating Latent Topologies for Implicit 3D Reconstructioncited by 21
  12. 2024RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptionscited by 21
26 more
  1. 2021Stereoscopic Universal Perturbations across Different Architectures and Datasetscited by 17
  2. 2023Enhancing Diffusion Models with 3D Perspective Geometry Constraintscited by 16
  3. 2023AugUndo: Scaling Up Augmentations for Monocular Depth Completion and Estimationcited by 15
  4. 2023Learnable Differencing Center for Nighttime Depth Perceptioncited by 14
  5. 2022Towards 3D Scene Reconstruction from Locally Scale-Aligned Monocular Video Depthcited by 11
  6. 2023SuperPrimitive: Scene Reconstruction at a Primitive Levelcited by 11
  7. 2025ProtoDepth: Unsupervised Continual Depth Completion with Prototypescited by 11
  8. 2022Struct-MDC: Mesh-Refined Unsupervised Depth Completion Leveraging Structural Regularities from Visual SLAMcited by 10
  9. 2025Extending Foundational Monocular Depth Estimators to Fisheye Cameras with Calibration Tokenscited by 9
  10. 2024RGB Guided ToF Imaging System: A Survey of Deep Learning-based Methodscited by 8
  11. 2024All-day Depth Completioncited by 8
  12. 2022MIPI 2022 Challenge on RGB+ToF Depth Completion: Dataset and Reportcited by 7
  13. 2025Zero-shot Depth Completion via Test-time Alignment with Affine-invariant Depth Priorcited by 6
  14. 2024UnCLe: Benchmarking Unsupervised Continual Learning for Depth Completioncited by 5
  15. 2025Radar-Guided Polynomial Fitting for Metric Depth Estimationcited by 5
  16. 2025ETA: Energy-based Test-time Adaptation for Depth Completioncited by 4
  17. 2022Sparse SPN: Depth Completion from Sparse Keypointscited by 3
  18. 2022Segmentation-guided Domain Adaptation for Efficient Depth Completioncited by 2
  19. 2024Deep Cost Ray Fusion for Sparse Depth Video Completioncited by 1
  20. 2025StarryGazer: Leveraging Monocular Depth Estimation Models for Domain-Agnostic Single Depth Image Completioncited by 1
  21. 2026VisTa3D: A Dataset and Benchmark for Thin Object Reconstruction from Vision, Tactile, and 3D Point Cloudscited by 1
  22. 2022On the Viability of Monocular Depth Pre-training for Semantic Segmentationcited by 0
  23. 2023DEUX: Active Exploration for Learning Unsupervised Depth Perceptioncited by 0
  24. 2023Self-Supervised Depth Completion Guided by 3D Perception and Geometry Consistencycited by 0
  25. 2025Training Self-Supervised Depth Completion Using Sparse Measurements and a Single Imagecited by 0
  26. 2026From Perspective to Fisheye Depth Estimation and Open-Vocabulary Segmentationcited by 0