{"id":18488365,"url":"https://github.com/choyingw/scadc-depthcompletion","last_synced_at":"2026-03-15T18:13:09.347Z","repository":{"id":112372206,"uuid":"353875328","full_name":"choyingw/SCADC-DepthCompletion","owner":"choyingw","description":"ICASSP 2021: Scene Completeness-Aware Lidar Depth Completion for Driving Scenario","archived":false,"fork":false,"pushed_at":"2022-06-14T19:44:15.000Z","size":45368,"stargazers_count":18,"open_issues_count":1,"forks_count":1,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-06-26T08:02:50.587Z","etag":null,"topics":["3d","autonomous-driving","autonomous-vehicles","computer-vision","deep-neural-networks","depth-completion","depth-estimation","icassp","icassp2021","lidar","scene-reconstruction","stereo-vision"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/choyingw.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2021-04-02T01:52:43.000Z","updated_at":"2025-03-21T09:09:09.000Z","dependencies_parsed_at":"2023-05-14T01:00:17.168Z","dependency_job_id":null,"html_url":"https://github.com/choyingw/SCADC-DepthCompletion","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/choyingw/SCADC-DepthCompletion","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/choyingw%2FSCADC-DepthCompletion","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/choyingw%2FSCADC-DepthCompletion/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/choyingw%2FSCADC-DepthCompletion/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/choyingw%2FSCADC-DepthCompletion/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/choyingw","download_url":"https://codeload.github.com/choyingw/SCADC-DepthCompletion/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/choyingw%2FSCADC-DepthCompletion/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":266586409,"owners_count":23952172,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-07-22T02:00:09.085Z","response_time":66,"last_error":null,"robots_txt_status":null,"robots_txt_updated_at":null,"robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["3d","autonomous-driving","autonomous-vehicles","computer-vision","deep-neural-networks","depth-completion","depth-estimation","icassp","icassp2021","lidar","scene-reconstruction","stereo-vision"],"created_at":"2024-11-06T12:51:37.113Z","updated_at":"2026-03-15T18:13:04.307Z","avatar_url":"https://github.com/choyingw.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# SCADC-DepthCompletion\nScene Completeness-Aware Lidar Depth Completion for Driving Scenario, ICASSP 2021\n\nCho-Ying Wu and Ulrich Neumann, University of Southern California\n\n\u003cimg src='demo.gif'\u003e\n\nThe full example video link is here https://www.youtube.com/watch?v=FQDTdpMPKxs\n\nPaper: https://arxiv.org/abs/2003.06945\n\nProject page: https://choyingw.github.io/works/SCADC/index.html\n\n**Advantages:**\n\n👍 **First research to attend scene-completeness in depth completion**\n\n👍 **Sensor Fusion for lidar and stereo cameras**\n\n👍 **Structured upper scene depth**\n\n👍 **Precise lower scene**\n\n# Prerequisite\n\n\tUbuntu 16.04/ 20.04\n\tPython 3\n\tPyTorch 1.5+ (Tested on 1.5, should be compatiable for following versions)\n\tNVIDIA GPU + CUDA CuDNN \n\tOther common libraries: matplotlib, cv2, PIL\n\n# Data Preparation\n\nClone the repo first.\n\nThen, download preprocessed data from \u003ca href=\"https://drive.google.com/file/d/1c78Ox6KfaUkXZf4qx5hVly9Na_QJ5VIv/view?usp=sharing\"\u003etrain\u003c/a\u003e (142G) \u003ca href=\"https://drive.google.com/file/d/1RXJ5GFhE0ZIIBf4wcLXhilu4OVQ1BiEg/view?usp=sharing\"\u003eval\u003c/a\u003e (11G). This data includes training/val split that follows KITTI Completion and all required pre-processed data for this work.\n\nExtract the files under the repository. The structure should be like 'SCADC-DepthCompletion/Data/train' and 'SCADC-DepthCompletion/Data/val'\n\n\\*.h5 files are provided, including sparse depth (D), semi-dense depth (D_semi), left-right pairs (I_L and I_R), depth completed from \u003ca href=\"https://github.com/fangchangma/self-supervised-depth-completion\"\u003eSSDC\u003c/a\u003e (depth_c), and disparity from \u003ca href=\"https://github.com/JiaRenChang/PSMNet\"\u003ePSMNet\u003c/a\u003e (disp_c).\n\n# Evaluation/Training Commands:\n\nOur provided pretrained weight is under './test_ckpt/kitti/'. To quickly get our scene completeness-aware depth maps, just use the evaluation command, and it will save frame-by-frame results under './vis/'. Download \"val\" data split in the Data Preparation section and unzip under 'data/'. The folder structure and the evaluation command should be\n\n      .\n      ├── data\n            ├── val\n               ├── 0\n                   ├── 00000.h5\n\t\t     ......\n\t\t     \n\tpython3 evaluate.py --name kitti --checkpoints_dir './test_ckpt' --test_path ./data\n\nThis is the training command is you want ot train the network yourself.\n\n\tpython3 train_depth_complete.py --name kitti --checkpoints_dir [preferred saving ckpt path] --train_path [train_data_dir] --test_path [test_data_dir]\n\n\\[train_data_dir\\]: it should be 'Data/train' when you follow the recommended folder structure\n\\[test_data_dir\\]: it should be 'Data/test' when you follow the recommended folder structure\n\n# Customized depth completion and stereo estimation base methods:\n\nNote that we use \u003ca href=\"https://github.com/fangchangma/self-supervised-depth-completion\"\u003eSSDC\u003c/a\u003e, and disparity from \u003ca href=\"https://github.com/JiaRenChang/PSMNet\"\u003ePSMNet\u003c/a\u003e. \n\nThe pre-processed data is in the \\*.h5 files. (key: 'depth_c' and 'disp_c'). If you want to make completion results from different basic methods, please prepare those data at your own and replace data stored in \\*.h5 files.\n\n\nIf you find our work useful, please consider to cite our work.\n\n\t@inproceedings{wu2021scene,\n\t  title={Scene Completeness-Aware Lidar Depth Completion for Driving Scenario},\n\t  author={Wu, Cho-Ying and Neumann, Ulrich},\n\t  booktitle={ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},\n\t  pages={2490--2494},\n\t  year={2021},\n\t  organization={IEEE}\n\t}\n\n\n# Acknowledgement\n\nThe code development is based on \u003ca href=\"https://github.com/choyingw/CFCNet\"\u003eCFCNet\u003c/a\u003e, \u003ca href=\"https://github.com/fangchangma/self-supervised-depth-completion\"\u003eSelf-Supervised Depth Completion\u003c/a\u003e, and \u003ca href=\"https://github.com/JiaRenChang/PSMNet\"\u003ePSMNet\u003c/a\u003e. \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchoyingw%2Fscadc-depthcompletion","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fchoyingw%2Fscadc-depthcompletion","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchoyingw%2Fscadc-depthcompletion/lists"}