{"id":13535331,"url":"https://github.com/MightyChaos/LKVOLearner","last_synced_at":"2025-04-02T01:30:28.524Z","repository":{"id":87286641,"uuid":"116078544","full_name":"MightyChaos/LKVOLearner","owner":"MightyChaos","description":"Learning Depth from Monocular Videos using Direct Methods, CVPR 2018","archived":false,"fork":false,"pushed_at":"2018-11-22T07:53:07.000Z","size":76,"stargazers_count":230,"open_issues_count":6,"forks_count":38,"subscribers_count":13,"default_branch":"master","last_synced_at":"2024-11-02T23:32:47.653Z","etag":null,"topics":["ddvo","depth-estimation","posenet-ddvo","pytorch","unsupervised-learning","visual-odometry"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/MightyChaos.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,"roadmap":null,"authors":null}},"created_at":"2018-01-03T01:59:18.000Z","updated_at":"2024-07-24T04:47:34.000Z","dependencies_parsed_at":"2023-05-04T05:47:24.601Z","dependency_job_id":null,"html_url":"https://github.com/MightyChaos/LKVOLearner","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MightyChaos%2FLKVOLearner","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MightyChaos%2FLKVOLearner/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MightyChaos%2FLKVOLearner/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MightyChaos%2FLKVOLearner/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/MightyChaos","download_url":"https://codeload.github.com/MightyChaos/LKVOLearner/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246738366,"owners_count":20825772,"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","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":["ddvo","depth-estimation","posenet-ddvo","pytorch","unsupervised-learning","visual-odometry"],"created_at":"2024-08-01T08:00:53.913Z","updated_at":"2025-04-02T01:30:28.236Z","avatar_url":"https://github.com/MightyChaos.png","language":"Python","funding_links":[],"categories":["2. Monocular Depth (Semi- / Un-Supervised)","2. 单目深度估计(半监督、无监督)"],"sub_categories":["2.2 Multi View"],"readme":"# Learning Depth from Monocular Videos using Direct Methods\n\u003cimg align=\"center\" src=\"https://github.com/MightyChaos/MightyChaos.github.io/blob/master/projects/cvpr18_chaoyang/demo.gif\"\u003e\n\nImplementation of the methods in \"[Learning Depth from Monocular Videos using Direct Methods](http://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_Learning_Depth_From_CVPR_2018_paper.pdf)\".\nIf you find this code useful, please cite our paper:\n\n```\n@InProceedings{Wang_2018_CVPR,\nauthor = {Wang, Chaoyang and Miguel Buenaposada, José and Zhu, Rui and Lucey, Simon},\ntitle = {Learning Depth From Monocular Videos Using Direct Methods},\nbooktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},\nmonth = {June},\nyear = {2018}\n}\n```\n## Dependencies\n- Python 3.6\n- PyTorch 0.3.1  (latter or eariler version of Pytorch is non-compatible.)\n\n- visdom, dominate \n\n\n## Training\n### data preparation\nWe refer \"[SfMLeaner](https://github.com/tinghuiz/SfMLearner)\" to prepare the training data from KITTI. We assume the processed data is put in directory \"./data_kitti/\".\n\n### training with different pose prediction modules\nStart visdom server before for inspecting learning progress before starting the training process.\n```\npython -m visdom.server -port 8009\n```\n1. #### train from scratch with PoseNet\n```\nbash run_train_posenet.sh\n```\nsee [run_train_posenet.sh](https://github.com/MightyChaos/LKVOLearner/blob/master/run_train_posenet.sh) for details.\n\n2. #### finetune with DDVO\nUse pretrained posenet to give initialization for DDVO. Corresponds to the results reported as \"PoseNet+DDVO\" in the paper.\n```\nbash run_train_finetune.sh\n```\nsee [run_train_finetune.sh](https://github.com/MightyChaos/LKVOLearner/blob/master/run_train_finetune.sh) for details.\n\n## Testing\n- Pretrained depth network reported as \"Posenet-DDVO(CS+K)\" in the paper [[download](https://drive.google.com/file/d/1SJWLfA7kqpERj_U2gYXl7Vuy1eQyOO_K/view?usp=sharing)].\n- Depth prediction results on KITTI eigen test split(see Table 1 in the paper):   [[Posenet(K)](https://drive.google.com/open?id=1Wj7ulSimrvrzNx4TRd-JspmX3DJwgPiV)], [[DDVO(K)](https://drive.google.com/open?id=1wiODwgX_Vm_w7fVK1y_X5CNJTtgaPwcN)], [[Posenet+DDVO(K)](https://drive.google.com/open?id=1uUQJLcUOoY2hG6QS_F-wbM3GDAjD-Z5h)],[[Posenet+DDVO(CS+K)](https://drive.google.com/open?id=1hp4zFgK5NSNGdvaQL2ZumeinMQY_-AwK)]\n\n- To test yourself:\n```\nCUDA_VISIBLE_DEVICES=0 nice -10 python src/testKITTI.py --dataset_root $DATAROOT --ckpt_file $CKPT --output_path $OUTPUT --test_file_list test_files_eigen.txt\n```\n\n## Evaluation\nWe again refer to \"[SfMLeaner](https://github.com/tinghuiz/SfMLearner)\" for their evaluation code.\n\n\n## Acknowledgement\nPart of the code structure is borrowed from \"[Pytorch CycleGAN](https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix)\"","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FMightyChaos%2FLKVOLearner","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FMightyChaos%2FLKVOLearner","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FMightyChaos%2FLKVOLearner/lists"}