{"url":"/task/depth-completion","name":"Depth Completion","slug":"depth-completion","description_markdown":"The **Depth Completion** task is a sub-problem of depth estimation. In the sparse-to-dense depth completion problem, one wants to infer the dense depth map of a 3-D scene given an RGB image and its corresponding sparse reconstruction in the form of a sparse depth map obtained either from computational methods such as SfM (Strcuture-from-Motion) or active sensors such as lidar or structured light sensors. \r\n\r\n<span class=\"description-source\">Source: [LiStereo: Generate Dense Depth Maps from LIDAR and Stereo Imagery ](https://arxiv.org/abs/1905.02744)</span>,\r\n<span class=\"description-source\">[Unsupervised Depth Completion from Visual Inertial Odometry](https://arxiv.org/abs/1905.08616)</span>","categories":[{"name":"Computer Vision","url":"/area/computer-vision"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":242,"papers_with_code":96,"benchmarks":9,"benchmark_tables_in_archive":9,"benchmark_tables_shown":9,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":11,"subtasks":0,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/depth-completion-on-kitti-depth-completion","slug":"depth-completion-on-kitti-depth-completion","dataset":"KITTI Depth Completion","dataset_url":null,"rows_in_archive":16,"metrics":["RMSE","MAE","iRMSE","iMAE","Runtime [ms]"],"first_row_in_archive_order":{"model":"SemAttNet","paper_title":"SemAttNet: Towards Attention-based Semantic Aware Guided Depth Completion","paper_url":"/paper/semattnet-towards-attention-based-semantic","paper_date":"2022-04-28","arxiv_id":"2204.13635","code_links":[{"title":"danishnazir/SemAttNet","url":"https://github.com/danishnazir/SemAttNet"}],"syntology":null}},{"leaderboard":"/sota/depth-completion-on-void","slug":"depth-completion-on-void","dataset":"VOID","dataset_url":"/dataset/void","rows_in_archive":6,"metrics":["MAE","RMSE","iMAE","iRMSE"],"first_row_in_archive_order":{"model":"NLSPN","paper_title":"Non-Local Spatial Propagation Network for Depth Completion","paper_url":"/paper/non-local-spatial-propagation-network-for","paper_date":"2020-07-20","arxiv_id":"2007.10042","code_links":[{"title":"zzangjinsun/NLSPN_ECCV20","url":"https://github.com/zzangjinsun/NLSPN_ECCV20"}],"syntology":{"n":5,"n_ran":1,"n_unverified":4,"n_pointer_only":0}}},{"leaderboard":"/sota/depth-completion-on-nyu-depth-v2","slug":"depth-completion-on-nyu-depth-v2","dataset":"NYU-Depth V2","dataset_url":"/dataset/nyuv2","rows_in_archive":3,"metrics":["RMSE","REL"],"first_row_in_archive_order":{"model":"NLSPN","paper_title":"Non-Local Spatial Propagation Network for Depth Completion","paper_url":"/paper/non-local-spatial-propagation-network-for","paper_date":"2020-07-20","arxiv_id":"2007.10042","code_links":[{"title":"zzangjinsun/NLSPN_ECCV20","url":"https://github.com/zzangjinsun/NLSPN_ECCV20"}],"syntology":{"n":5,"n_ran":1,"n_unverified":4,"n_pointer_only":0}}},{"leaderboard":"/sota/depth-completion-on-matterport3d","slug":"depth-completion-on-matterport3d","dataset":"Matterport3D","dataset_url":"/dataset/matterport3d","rows_in_archive":2,"metrics":["RMSE"],"first_row_in_archive_order":{"model":"DM-LRN-b4","paper_title":"Decoder Modulation for Indoor Depth Completion","paper_url":"/paper/decoder-modulation-for-indoor-depth","paper_date":"2020-05-18","arxiv_id":"2005.08607","code_links":[],"syntology":null}},{"leaderboard":"/sota/depth-completion-on-kitti","slug":"depth-completion-on-kitti","dataset":"KITTI","dataset_url":"/dataset/kitti","rows_in_archive":1,"metrics":["RMSE"],"first_row_in_archive_order":{"model":"FusionDepth","paper_title":"Advancing Self-supervised Monocular Depth Learning with Sparse LiDAR","paper_url":"/paper/advancing-self-supervised-monocular-depth","paper_date":"2021-09-20","arxiv_id":"2109.09628","code_links":[{"title":"AutoAILab/FusionDepth","url":"https://github.com/AutoAILab/FusionDepth"},{"title":"fengziyue/FusionDepth","url":"https://github.com/fengziyue/FusionDepth"}],"syntology":{"n":15,"n_ran":4,"n_unverified":11,"n_pointer_only":0}}},{"leaderboard":"/sota/depth-completion-on-kitti-depth-completion-1","slug":"depth-completion-on-kitti-depth-completion-1","dataset":"KITTI Depth Completion Eigen Split","dataset_url":null,"rows_in_archive":1,"metrics":["REL","RMSE"],"first_row_in_archive_order":{"model":"DSN","paper_title":"On Deep Learning Techniques to Boost Monocular Depth Estimation for Autonomous Navigation","paper_url":"/paper/on-deep-learning-techniques-to-boost","paper_date":"2020-10-13","arxiv_id":"2010.06626","code_links":[],"syntology":null}},{"leaderboard":"/sota/depth-completion-on-kitti-depth-completion-2","slug":"depth-completion-on-kitti-depth-completion-2","dataset":"KITTI Depth Completion 500 points","dataset_url":null,"rows_in_archive":1,"metrics":["RMSE "],"first_row_in_archive_order":{"model":"CFCNet","paper_title":"Deep RGB-D Canonical Correlation Analysis For Sparse Depth Completion","paper_url":"/paper/deep-rgb-d-canonical-correlation-analysis-for","paper_date":"2019-06-21","arxiv_id":"1906.08967","code_links":[{"title":"choyingw/CFCNet","url":"https://github.com/choyingw/CFCNet"}],"syntology":null}},{"leaderboard":"/sota/depth-completion-on-plad","slug":"depth-completion-on-plad","dataset":"PLAD","dataset_url":"/dataset/plad-1","rows_in_archive":1,"metrics":["MAE","RMSE"],"first_row_in_archive_order":{"model":"Struct-MDC","paper_title":"Struct-MDC: Mesh-Refined Unsupervised Depth Completion Leveraging Structural Regularities from Visual SLAM","paper_url":"/paper/struct-mdc-mesh-refined-unsupervised-depth","paper_date":"2022-04-29","arxiv_id":"2204.13877","code_links":[{"title":"url-kaist/Struct-MDC","url":"https://github.com/url-kaist/Struct-MDC"}],"syntology":null}},{"leaderboard":"/sota/depth-completion-on-void-150","slug":"depth-completion-on-void-150","dataset":"VOID-150","dataset_url":null,"rows_in_archive":1,"metrics":["MAE","RMSE"],"first_row_in_archive_order":{"model":"Struct-MDC","paper_title":"Struct-MDC: Mesh-Refined Unsupervised Depth Completion Leveraging Structural Regularities from Visual SLAM","paper_url":"/paper/struct-mdc-mesh-refined-unsupervised-depth","paper_date":"2022-04-29","arxiv_id":"2204.13877","code_links":[{"title":"url-kaist/Struct-MDC","url":"https://github.com/url-kaist/Struct-MDC"}],"syntology":null}}],"datasets":[{"url":"/dataset/kitti","name":"KITTI","full_name":"","num_papers_in_archive":3661},{"url":"/dataset/nyuv2","name":"NYUv2","full_name":"NYU-Depth V2","num_papers_in_archive":986},{"url":"/dataset/matterport3d","name":"Matterport3D","full_name":"","num_papers_in_archive":461},{"url":"/dataset/void","name":"VOID","full_name":"Visual Odometry with Inertial and Depth","num_papers_in_archive":26},{"url":"/dataset/kitti-depth","name":"KITTI-Depth","full_name":null,"num_papers_in_archive":14},{"url":"/dataset/transcg","name":"TransCG","full_name":"TransCG","num_papers_in_archive":7},{"url":"/dataset/supercaustics","name":"SuperCaustics","full_name":"","num_papers_in_archive":2},{"url":"/dataset/bidcd","name":"BIDCD","full_name":"Bosch Industrial Depth Completion Dataset","num_papers_in_archive":1},{"url":"/dataset/conslam","name":"ConSLAM","full_name":"Construction Dataset for SLAM","num_papers_in_archive":1},{"url":"/dataset/plad-1","name":"PLAD","full_name":"Point Line and Depth dataset","num_papers_in_archive":1},{"url":"/dataset/ms2-dataset-rgb-nir-thermal-images-lidar-gps","name":"Multi-Spectral Stereo Dataset  (RGB, NIR, thermal images, LiDAR, GPS/IMU)","full_name":"","num_papers_in_archive":0}],"subtasks":[],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":96,"tagged_in_all":242,"items":[{"url":"/paper/parse-geometry-from-a-line-monocular-depth","title":"Parse Geometry from a Line: Monocular Depth Estimation with Partial Laser Observation","date":"2016-10-17","arxiv_id":"1611.02174","repositories_listed":4,"syntology":null},{"url":"/paper/towards-precise-and-efficient-image-guided","title":"PENet: Towards Precise and Efficient Image Guided Depth Completion","date":"2021-03-01","arxiv_id":"2103.00783","repositories_listed":3,"syntology":null},{"url":"/paper/indoor-depth-completion-with-boundary","title":"Indoor Depth Completion with Boundary Consistency and Self-Attention","date":"2019-08-22","arxiv_id":"1908.08344","repositories_listed":3,"syntology":{"n":3,"n_ran":0,"n_unverified":3,"n_pointer_only":0}},{"url":"/paper/noise-aware-unsupervised-deep-lidar-stereo","title":"Noise-Aware Unsupervised Deep Lidar-Stereo Fusion","date":"2019-04-08","arxiv_id":"1904.03868","repositories_listed":3,"syntology":null},{"url":"/paper/cfpnet-improving-lightweight-tof-depth","title":"CFPNet: Improving Lightweight ToF Depth Completion via Cross-zone Feature Propagation","date":"2024-11-07","arxiv_id":"2411.04480","repositories_listed":2,"syntology":null},{"url":"/paper/a-concise-but-effective-network-for-image","title":"A Concise but High-performing Network for Image Guided Depth Completion in Autonomous Driving","date":"2024-01-29","arxiv_id":"2401.15902","repositories_listed":2,"syntology":null},{"url":"/paper/revisiting-deformable-convolution-for-depth","title":"Revisiting Deformable Convolution for Depth Completion","date":"2023-08-03","arxiv_id":"2308.01905","repositories_listed":2,"syntology":null},{"url":"/paper/advancing-self-supervised-monocular-depth","title":"Advancing Self-supervised Monocular Depth Learning with Sparse LiDAR","date":"2021-09-20","arxiv_id":"2109.09628","repositories_listed":2,"syntology":{"n":15,"n_ran":4,"n_unverified":11,"n_pointer_only":0}},{"url":"/paper/aerial-single-view-depth-completion-with","title":"Aerial Single-View Depth Completion with Image-Guided Uncertainty Estimation","date":"2020-01-17","arxiv_id":null,"repositories_listed":2,"syntology":null},{"url":"/paper/learning-guided-convolutional-network-for","title":"Learning Guided Convolutional Network for Depth Completion","date":"2019-08-03","arxiv_id":"1908.01238","repositories_listed":2,"syntology":null},{"url":"/paper/190508616","title":"Unsupervised Depth Completion from Visual Inertial Odometry","date":"2019-05-15","arxiv_id":"1905.08616","repositories_listed":2,"syntology":null},{"url":"/paper/self-supervised-sparse-to-dense-self","title":"Self-supervised Sparse-to-Dense: Self-supervised Depth Completion from LiDAR and Monocular Camera","date":"2018-07-01","arxiv_id":"1807.00275","repositories_listed":2,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/in-defense-of-classical-image-processing-fast","title":"In Defense of Classical Image Processing: Fast Depth Completion on the CPU","date":"2018-01-31","arxiv_id":"1802.00036","repositories_listed":2,"syntology":null},{"url":"/paper/pacgdc-label-efficient-generalizable-depth","title":"PacGDC: Label-Efficient Generalizable Depth Completion with Projection Ambiguity and Consistency","date":"2025-07-10","arxiv_id":"2507.07374","repositories_listed":1,"syntology":null},{"url":"/paper/depthor-depth-enhancement-from-a-practical","title":"DEPTHOR: Depth Enhancement from a Practical Light-Weight dToF Sensor and RGB Image","date":"2025-04-02","arxiv_id":"2504.01596","repositories_listed":1,"syntology":null},{"url":"/paper/vanishing-depth-a-depth-adapter-with","title":"Vanishing Depth: A Depth Adapter with Positional Depth Encoding for Generalized Image Encoders","date":"2025-03-25","arxiv_id":"2503.19947","repositories_listed":1,"syntology":null},{"url":"/paper/svdc-consistent-direct-time-of-flight-video","title":"SVDC: Consistent Direct Time-of-Flight Video Depth Completion with Frequency Selective Fusion","date":"2025-03-03","arxiv_id":"2503.01257","repositories_listed":1,"syntology":{"n":14,"n_ran":2,"n_unverified":12,"n_pointer_only":0}},{"url":"/paper/transplat-surface-embedding-guided-3d","title":"TranSplat: Surface Embedding-guided 3D Gaussian Splatting for Transparent Object Manipulation","date":"2025-02-11","arxiv_id":"2502.07840","repositories_listed":1,"syntology":null},{"url":"/paper/tdcnet-transparent-objects-depth-completion","title":"TDCNet: Transparent Objects Depth Completion with CNN-Transformer Dual-Branch Parallel Network","date":"2024-12-19","arxiv_id":"2412.14961","repositories_listed":1,"syntology":null},{"url":"/paper/omni-dc-highly-robust-depth-completion-with","title":"OMNI-DC: Highly Robust Depth Completion with Multiresolution Depth Integration","date":"2024-11-28","arxiv_id":"2411.19278","repositories_listed":1,"syntology":null},{"url":"/paper/helvipad-a-real-world-dataset-for","title":"Helvipad: A Real-World Dataset for Omnidirectional Stereo Depth Estimation","date":"2024-11-27","arxiv_id":"2411.18335","repositories_listed":1,"syntology":null},{"url":"/paper/scale-propagation-network-for-generalizable","title":"Scale Propagation Network for Generalizable Depth Completion","date":"2024-10-24","arxiv_id":"2410.18408","repositories_listed":1,"syntology":null},{"url":"/paper/steeredmarigold-steering-diffusion-towards","title":"SteeredMarigold: Steering Diffusion Towards Depth Completion of Largely Incomplete Depth Maps","date":"2024-09-16","arxiv_id":"2409.10202","repositories_listed":1,"syntology":null},{"url":"/paper/sdformer-efficient-end-to-end-transformer-for","title":"SDformer: Efficient End-to-End Transformer for Depth Completion","date":"2024-09-12","arxiv_id":"2409.08159","repositories_listed":1,"syntology":null},{"url":"/paper/ogni-dc-robust-depth-completion-with","title":"OGNI-DC: Robust Depth Completion with Optimization-Guided Neural Iterations","date":"2024-06-17","arxiv_id":"2406.11711","repositories_listed":1,"syntology":{"n":11,"n_ran":9,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/a-two-stage-masked-autoencoder-based-network","title":"A Two-Stage Masked Autoencoder Based Network for Indoor Depth Completion","date":"2024-06-14","arxiv_id":"2406.09792","repositories_listed":1,"syntology":null},{"url":"/paper/masked-spatial-propagation-network-for","title":"Masked Spatial Propagation Network for Sparsity-Adaptive Depth Refinement","date":"2024-04-30","arxiv_id":"2404.19294","repositories_listed":1,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/roofdiffusion-constructing-roofs-from","title":"RoofDiffusion: Constructing Roofs from Severely Corrupted Point Data via Diffusion","date":"2024-04-14","arxiv_id":"2404.09290","repositories_listed":1,"syntology":{"n":2,"n_ran":0,"n_unverified":2,"n_pointer_only":2}},{"url":"/paper/neslam-neural-implicit-mapping-and-self","title":"NeSLAM: Neural Implicit Mapping and Self-Supervised Feature Tracking With Depth Completion and Denoising","date":"2024-03-29","arxiv_id":"2403.20034","repositories_listed":1,"syntology":null},{"url":"/paper/bilateral-propagation-network-for-depth","title":"Bilateral Propagation Network for Depth Completion","date":"2024-03-17","arxiv_id":"2403.11270","repositories_listed":1,"syntology":{"n":6,"n_ran":1,"n_unverified":5,"n_pointer_only":0}}],"syntology_records":8,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}