{"id":13791278,"url":"https://github.com/FrozenBurning/Relighting4D","last_synced_at":"2025-05-12T10:31:31.504Z","repository":{"id":45518924,"uuid":"512430552","full_name":"FrozenBurning/Relighting4D","owner":"FrozenBurning","description":"[ECCV 2022] Relighting4D: Neural Relightable Human from Videos","archived":false,"fork":false,"pushed_at":"2023-02-16T02:23:32.000Z","size":223,"stargazers_count":268,"open_issues_count":3,"forks_count":19,"subscribers_count":6,"default_branch":"master","last_synced_at":"2024-11-18T05:38:51.356Z","etag":null,"topics":["eccv2022","illumination","nerf","neural-rendering","reflectance","relighting","view-synthesis"],"latest_commit_sha":null,"homepage":"https://frozenburning.github.io/projects/relighting4d/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/FrozenBurning.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":"2022-07-10T12:34:58.000Z","updated_at":"2024-11-06T20:44:28.000Z","dependencies_parsed_at":"2024-01-07T05:37:37.045Z","dependency_job_id":null,"html_url":"https://github.com/FrozenBurning/Relighting4D","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/FrozenBurning%2FRelighting4D","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/FrozenBurning%2FRelighting4D/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/FrozenBurning%2FRelighting4D/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/FrozenBurning%2FRelighting4D/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/FrozenBurning","download_url":"https://codeload.github.com/FrozenBurning/Relighting4D/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253719936,"owners_count":21952923,"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":["eccv2022","illumination","nerf","neural-rendering","reflectance","relighting","view-synthesis"],"created_at":"2024-08-03T22:00:58.212Z","updated_at":"2025-05-12T10:31:31.118Z","avatar_url":"https://github.com/FrozenBurning.png","language":"Python","funding_links":[],"categories":["3d human"],"sub_categories":["nerf_pifu_3dgs"],"readme":"\u003cdiv align=\"center\"\u003e\n\n\u003ch1\u003eRelighting4D: Neural Relightable Human from Videos\u003c/h1\u003e\n\n\u003cdiv\u003e\n    \u003ca href='https://frozenburning.github.io/' target='_blank'\u003eZhaoxi Chen\u003c/a\u003e\u0026emsp;\n    \u003ca href='https://liuziwei7.github.io/' target='_blank'\u003eZiwei Liu\u003c/a\u003e\n\u003c/div\u003e\n\u003cdiv\u003e\n    S-Lab, Nanyang Technological University\n\u003c/div\u003e\n\n\u003cstrong\u003e\u003ca href='https://eccv2022.ecva.net/' target='_blank'\u003eECCV 2022\u003c/a\u003e\u003c/strong\u003e\n\n### [Project Page](https://frozenburning.github.io/projects/relighting4d) | [Video](https://youtu.be/NayAw89qtsY) | [Paper](https://arxiv.org/abs/2207.07104)\n\n\u003ctr\u003e\n    \u003cimg src=\"https://github.com/FrozenBurning/FrozenBurning.github.io/blob/master/projects/relighting4d/img/teaser.gif\" width=\"100%\"/\u003e\n\u003c/tr\u003e\n\u003c/div\u003e\n\n## Updates\n\n[08/2022] Model weights released. [![Google Drive](https://img.shields.io/badge/Google%20Drive-4285F4?style=for-the-badge\u0026logo=googledrive\u0026logoColor=yellow)](https://drive.google.com/drive/folders/14pvUxVNCrKEFYjy3h2nc_i1Y7rrcKq2q?usp=sharing)\n\n[07/2022] Paper uploaded to arXiv. [![arXiv](https://img.shields.io/badge/arXiv-2207.07104-b31b1b.svg)](https://arxiv.org/abs/2207.07104)\n\n[07/2022] Code released.\n\n## Citation\n\nIf you find our work useful for your research, please consider citing this paper:\n\n```\n@inproceedings{chen2022relighting,\n    title={Relighting4D: Neural Relightable Human from Videos},\n    author={Zhaoxi Chen and Ziwei Liu},\n    booktitle={ECCV},\n    year={2022}\n}\n```\n\n\n## Installation\nWe recommend using [Anaconda](https://www.anaconda.com/) to manage your python environment. You can setup the required environment by the following command:\n```(bash)\nconda env create -f environment.yml\nconda activate relighting4d\n```\n\n## Datasets\n\n### People-Snapshot\nWe follow [NeuralBody](https://github.com/zju3dv/neuralbody) for data preparation.\n1. Download the People-Snapshot dataset [here](https://graphics.tu-bs.de/people-snapshot).\n2. Process the People-Snapshot dataset using the [script](tools/process_snapshot.py).\n3. Create a soft link:\n\n    ```(bash)\n    cd /path/to/Relighting4D\n    mkdir -p data\n    cd data\n    ln -s /path/to/people_snapshot people_snapshot\n    ```\n\n### ZJU-MoCap\nPlease refer to [here](https://github.com/zju3dv/neuralbody/blob/master/INSTALL.md) for requesting the download link. Once downloaded, don't forget to add a soft link:\n```(bash)\ncd /path/to/Relighting4D\nmkdir -p data\ncd data\nln -s /path/to/zju_mocap zju_mocap\n```\n\n## Training\nWe first reconstruct an auxiliary density field in Stage I and then train the whole pipeline in Stage II. All trainings are done on a Tesla V100 GPU with 16GB memory.\n\nTake the training on `female-3-casual` as an example.\n\n* Stage I:\n    ```(bash)\n    python train_net.py --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c resume False gpus \"0,\"\n    ```\n    The model weights will be saved to `/data/trained_model/if_nerf/female3c/latest.pth`.\n\n* Stage II:\n    ```(bash)\n    python train_net.py --cfg_file configs/snapshot_exp/snapshot_f3c.yaml task relighte2e exp_name female3c_relight train_relight True resume False train_relight_cfg.smpl_model_ckpt ./data/trained_model/if_nerf/female3c/latest.pth gpus \"0,\"\n    ```\n    The final model will be saved to `/data/trained_model/relighte2e/female3c_relight/latest.pth`.\n\n* Tensorboard:\n    ```\n    tensorboard --logdir data/record/if_nerf\n    tensorboard --logdir data/record/relighte2e\n    ```\n\n\n## Rendering\nTo relight a human performer from the trained video, our model requires an HDR environment map as input. We provide 8 HDR maps at [light-probes](light-probes/). You can also use your own HDRIs or download some samples from [Poly Haven](https://polyhaven.com/hdris).\n\nYou are welcome to download our checkpoints from [Google Drive](https://drive.google.com/drive/folders/14pvUxVNCrKEFYjy3h2nc_i1Y7rrcKq2q?usp=sharing). \n\nHere, we take the rendering on `female-3-casual` as an example. \n\n* Relight with novel views of single frame\n    ```(bash)\n    python run.py --type relight --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c_relight task relighte2e vis_relight True ratio 0.5 gpus \"0,\"\n    ```\n\n* Relight the dynamic humans in video frames\n    ```(bash)\n    python run.py --type relight_npose --cfg_file configs/snapshot_exp/snapshot_f3c.yaml exp_name female3c_relight task relighte2e vis_relight True vis_relight_npose True ratio 0.5 pyramid False gpus \"0,\"\n    ```\nThe results of rendering are located at `/data/render/`. For example, rendering results with [courtyard HDR environment](light-probes/courtyard.hdr) are shown as follows:\n\n\u003ctable\u003e\n\u003ctr\u003e\n    \u003ctd align='center' width='50%'\u003e\u003cimg src=\"https://frozenburning.github.io/projects/relighting4d/img/nview.gif\" width=\"100%\"/\u003e\u003c/td\u003e\n    \u003ctd align='center' width='50%'\u003e\u003cimg src=\"https://frozenburning.github.io/projects/relighting4d/img/npose.gif\" width=\"100%\"/\u003e\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/table\u003e\n\n## Acknowledgements\nThis work is supported by the National Research Foundation, Singapore under its AI Singapore Programme, NTU NAP, MOE AcRF Tier 2 (T2EP20221-0033), and under the RIE2020 Industry Alignment Fund - Industry Collaboration Projects (IAF-ICP) Funding Initiative, as well as cash and in-kind contribution from the industry partner(s).\n\nRelighting4D is implemented on top of the [NeuralBody](https://github.com/zju3dv/neuralbody) codebase.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FFrozenBurning%2FRelighting4D","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FFrozenBurning%2FRelighting4D","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FFrozenBurning%2FRelighting4D/lists"}