mlsanityData to 7 Oct 2026

In Defense of Classical Image Processing: Fast Depth Completion on the CPU

arXiv 1802.00036 ↗First posted 31 Jan 2018cs.CVcited by 96

Not on the map. Why?

Abstract (arXiv)

With the rise of data driven deep neural networks as a realization of universal function approximators, most research on computer vision problems has moved away from hand crafted classical image processing algorithms. This paper shows that with a well designed algorithm, we are capable of outperforming neural network based methods on the task of depth completion. The proposed algorithm is simple and fast, runs on the CPU, and relies only on basic image processing operations to perform depth completion of sparse LIDAR depth data. We evaluate our algorithm on the challenging KITTI depth completion benchmark, and at the time of submission, our method ranks first on the KITTI test server among all published methods. Furthermore, our algorithm is data independent, requiring no training data to perform the task at hand. The code written in Python will be made publicly available at https://github.com/kujason/ip_basic.

Key facts. Cited by 96 papers in the map (data through 7 Oct 2026). First posted on arXiv 31 Jan 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 18; 2022: 20; 2023: 18; 2024: 20; 2025: 11; 2026: 9.

96 papers.

Sort:

By area:

  1. 2021Categorical Depth Distribution Network for Monocular 3D Object Detectioncited by 326
  2. 2023OpenOccupancy: A Large Scale Benchmark for Surrounding Semantic Occupancy Perceptioncited by 218
  3. 2021FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimationcited by 144
  4. 2022DeepInteraction: 3D Object Detection via Modality Interactioncited by 125
  5. 2023S-NeRF: Neural Radiance Fields for Street Viewscited by 116
  6. 2021RigNet: Repetitive Image Guided Network for Depth Completioncited by 80
  7. 2022Dynamic Spatial Propagation Network for Depth Completioncited by 70
  8. 2022MonoDistill: Learning Spatial Features for Monocular 3D Object Detectioncited by 68
  9. 20223D Object Detection from Images for Autonomous Driving: A Surveycited by 66
  10. 2021Unsupervised Depth Completion with Calibrated Backprojection Layerscited by 61
  11. 2024Bilateral Propagation Network for Depth Completioncited by 59
  12. 2023LRRU: Long-short Range Recurrent Updating Networks for Depth Completioncited by 48
38 more
  1. 2024Marigold-DC: Zero-Shot Monocular Depth Completion with Guided Diffusioncited by 47
  2. 2022ONCE-3DLanes: Building Monocular 3D Lane Detectioncited by 44
  3. 2024Tri-Perspective View Decomposition for Geometry-Aware Depth Completioncited by 42
  4. 2022SuperFusion: Multilevel LiDAR-Camera Fusion for Long-Range HD Map Generationcited by 39
  5. 2021Learning Topology from Synthetic Data for Unsupervised Depth Completioncited by 38
  6. 2021Learning a State Representation and Navigation in Cluttered and Dynamic Environmentscited by 30
  7. 2021An Adaptive Framework for Learning Unsupervised Depth Completioncited by 30
  8. 2023FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimatorscited by 29
  9. 2021ILabel: Interactive Neural Scene Labellingcited by 28
  10. 2024S-NeRF++: Autonomous Driving Simulation via Neural Reconstruction and Generationcited by 28
  11. 2022DesNet: Decomposed Scale-Consistent Network for Unsupervised Depth Completioncited by 27
  12. 2024SPIN: Simultaneous Perception, Interaction and Navigationcited by 27
  13. 2022Monitored Distillation for Positive Congruent Depth Completioncited by 24
  14. 2021A Surface Geometry Model for LiDAR Depth Completioncited by 18
  15. 2023Revisiting Depth Completion from a Stereo Matching Perspective for Cross-domain Generalizationcited by 18
  16. 2021DenseLiDAR: A Real-Time Pseudo Dense Depth Guided Depth Completion Networkcited by 17
  17. 2022Improving Monocular Visual Odometry Using Learned Depthcited by 17
  18. 2021Robust Depth Completion with Uncertainty-Driven Loss Functionscited by 16
  19. 2023Deep Fusion Transformer Network with Weighted Vector-Wise Keypoints Voting for Robust 6D Object Pose Estimationcited by 16
  20. 20223DPPE: 3D Point Positional Encoding for Multi-Camera 3D Object Detection Transformerscited by 15
  21. 2024DeepInteraction++: Multi-Modality Interaction for Autonomous Drivingcited by 15
  22. 2023PointMBF: A Multi-scale Bidirectional Fusion Network for Unsupervised RGB-D Point Cloud Registrationcited by 14
  23. 2025GaussianFormer3D: Multi-Modal Gaussian-based Semantic Occupancy Prediction with 3D Deformable Attentioncited by 14
  24. 2021Radar Voxel Fusion for 3D Object Detectioncited by 13
  25. 2024Completion as Enhancement: A Degradation-Aware Selective Image Guided Network for Depth Completioncited by 12
  26. 2023RigNet++: Semantic Assisted Repetitive Image Guided Network for Depth Completioncited by 11
  27. 2025SAMFusion: Sensor-Adaptive Multimodal Fusion for 3D Object Detection in Adverse Weathercited by 10
  28. 2024WildOcc: A Benchmark for Off-Road 3D Semantic Occupancy Predictioncited by 9
  29. 2023You Only Look at One: Category-Level Object Representations for Pose Estimation From a Single Examplecited by 8
  30. 2025Self-Supervised Sparse Sensor Fusion for Long Range Perceptioncited by 8
  31. 2022M2-3DLaneNet: Exploring Multi-Modal 3D Lane Detectioncited by 7
  32. 2023Revisiting Fully Convolutional Geometric Features for Object 6D Pose Estimationcited by 7
  33. 2024SteeredMarigold: Steering Diffusion Towards Depth Completion of Largely Incomplete Depth Mapscited by 7
  34. 2022A Low Memory Footprint Quantized Neural Network for Depth Completion of Very Sparse Time-of-Flight Depth Mapscited by 6
  35. 2022CU-Net: LiDAR Depth-Only Completion With Coupled U-Netcited by 6
  36. 2021PDC: Piecewise Depth Completion utilizing Superpixelscited by 5
  37. 20233D Neural Embedding Likelihood: Probabilistic Inverse Graphics for Robust 6D Pose Estimationcited by 5
  38. 2023Learning to Estimate Two Dense Depths from LiDAR and Event Datacited by 5

50 most cited of 96 All 96 (JSON)