{"id":18614473,"url":"https://github.com/cgtuebingen/NeRD-Neural-Reflectance-Decomposition","last_synced_at":"2025-04-11T00:30:44.451Z","repository":{"id":43958882,"uuid":"319388912","full_name":"cgtuebingen/NeRD-Neural-Reflectance-Decomposition","owner":"cgtuebingen","description":"NeRD: Neural Reflectance Decomposition from Image Collections - ICCV 2021","archived":false,"fork":false,"pushed_at":"2023-04-08T17:37:44.000Z","size":293,"stargazers_count":250,"open_issues_count":6,"forks_count":26,"subscribers_count":25,"default_branch":"master","last_synced_at":"2024-11-07T03:31:42.260Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"https://markboss.me/publication/2021-nerd/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/cgtuebingen.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,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2020-12-07T17:07:31.000Z","updated_at":"2024-10-08T11:23:18.000Z","dependencies_parsed_at":"2024-11-07T03:40:38.166Z","dependency_job_id":null,"html_url":"https://github.com/cgtuebingen/NeRD-Neural-Reflectance-Decomposition","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/cgtuebingen%2FNeRD-Neural-Reflectance-Decomposition","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cgtuebingen%2FNeRD-Neural-Reflectance-Decomposition/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cgtuebingen%2FNeRD-Neural-Reflectance-Decomposition/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cgtuebingen%2FNeRD-Neural-Reflectance-Decomposition/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/cgtuebingen","download_url":"https://codeload.github.com/cgtuebingen/NeRD-Neural-Reflectance-Decomposition/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248322220,"owners_count":21084333,"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":[],"created_at":"2024-11-07T03:25:57.384Z","updated_at":"2025-04-11T00:30:43.865Z","avatar_url":"https://github.com/cgtuebingen.png","language":"Python","funding_links":[],"categories":["Papers"],"sub_categories":["NeRF Related Tasks"],"readme":"# NeRD: Neural Reflectance Decomposition from Image Collections\n\n### [Project Page](https://markboss.me/publication/2021-nerd/) | [Video](https://youtu.be/JL-qMTXw9VU) | [Paper](https://arxiv.org/abs/2012.03918) | [Dataset](download_datasets.py)\n\nImplementation for NeRD. A novel method which decomposes multiple images into shape, BRDF and illumination.\n\u003cbr\u003e\u003cbr\u003e\n[NeRD: Neural Reflectance Decomposition from Image Collections](https://markboss.me/publication/2021-nerd/)\u003cbr\u003e\n[Mark Boss](https://markboss.me)\u003csup\u003e1\u003c/sup\u003e, [Raphael Braun](https://uni-tuebingen.de/en/fakultaeten/mathematisch-naturwissenschaftliche-fakultaet/fachbereiche/informatik/lehrstuehle/computergrafik/lehrstuhl/mitarbeiter/raphael-braun/)\u003csup\u003e1\u003c/sup\u003e, [Varun Jampani](https://varunjampani.github.io)\u003csup\u003e2\u003c/sup\u003e, [Jonathan T. Barron](https://jonbarron.info)\u003csup\u003e2\u003c/sup\u003e, [Ce Liu](http://people.csail.mit.edu/celiu/)\u003csup\u003e2\u003c/sup\u003e, [Hendrik P. A. Lensch](https://uni-tuebingen.de/en/faculties/faculty-of-science/departments/computer-science/lehrstuehle/computergrafik/computer-graphics/staff/prof-dr-ing-hendrik-lensch/)\u003csup\u003e1\u003c/sup\u003e\u003cbr\u003e\n\u003csup\u003e1\u003c/sup\u003eUniversity of Tübingen, \u003csup\u003e2\u003c/sup\u003eGoogle Research \n\u003cbr\u003e\u003cbr\u003e\n![](images/teaser.jpg)\n\n**Also check our follow-up work: [Neural-PIL](https://github.com/cgtuebingen/Neural-PIL)**\n\n## Setup\n\nA conda environment is used for dependency management\n\n```\nconda env create -f environment.yml\nconda activate nerd\n```\n## Running\n\nReplace the specific `[]` placeholders:\n\n```\npython train_nerd.py --datadir [DIR_TO_DATASET_FOLDER] --basedir [TRAIN_DIR] --expname [EXPERIMENT_NAME] --gpu [COMMA_SEPARATED_GPU_LIST]\n```\n\n### Specific Arguments per Dataset\n\nMost setup is handled by configurations files located in [configs/nerd/](configs/nerd/).\n#### Our Synthethic Scenes\n\n```\n--config configs/nerd/blender.txt\n```\n\n#### NeRF Synthethic Scenes\n\n```\n--config configs/nerd/nerf_blender.txt\n```\n\n#### Real-World\n\n```\n--config configs/nerd/real_world.txt \n```\n\nOften objects are captured in a spherical manner and the flag `--spherify` should be applied for those scenes.\n\n## Datasets\n\nAll datasets are uploaded in individual git repositories. We have created a [download script](download_datasets.py) which automatically fetches all datasets and downloads them to a specified folder. Usage: \n\n```shell\npython download_datasets.py /path/to/dataset/root\n```\n## Run Your Own Data\n\nMainly camera poses and segmentation masks are required. For the poses the scripts from [NeRF](https://github.com/bmild/nerf#generating-poses-for-your-own-scenes) are used to prepare the scenes. The dataset then needs to be put in the following file structure:\n\n```\nimages/\n    [IMG_NAME_1].jpg\n    [IMG_NAME_2].jpg\n    ...\nmasks/\n    [IMG_NAME_1].jpg\n    [IMG_NAME_2].jpg\n    ...\nposes_bounds.npy\n```\n\nThe `poses_bounds.npy` is generated from the [LLFF script](https://github.com/bmild/nerf#dont-have-poses).\n\n## Evaluation\n\nThe [train_nerd.py](train_nerd.py) can be called with a `--render_only` flag and the `--config` flag pointing to the `args.txt` of the experiments folder. The SGs can be converted to regular environment maps with the `visualize_fit` function in [nn_utils/sg_rendering.py](nn_utils/sg_rendering.py) (Example in [train_nerd.py](train_nerd.py#L631)).\n## Citation\n\n```\n@inproceedings{boss2021nerd,\n  title         = {NeRD: Neural Reflectance Decomposition from Image Collections},\n  author        = {Boss, Mark and Braun, Raphael and Jampani, Varun and Barron, Jonathan T. and Liu, Ce and Lensch, Hendrik P.A.},\n  booktitle     = {IEEE International Conference on Computer Vision (ICCV)},\n  year          = {2021},\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcgtuebingen%2FNeRD-Neural-Reflectance-Decomposition","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcgtuebingen%2FNeRD-Neural-Reflectance-Decomposition","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcgtuebingen%2FNeRD-Neural-Reflectance-Decomposition/lists"}