{"id":13445083,"url":"https://apchenstu.github.io/TensoRF/","last_synced_at":"2025-03-20T19:31:31.000Z","repository":{"id":37633021,"uuid":"471096190","full_name":"apchenstu/TensoRF","owner":"apchenstu","description":"[ECCV 2022] Tensorial Radiance Fields, a novel approach to model and reconstruct radiance fields","archived":false,"fork":false,"pushed_at":"2023-09-27T06:08:20.000Z","size":33,"stargazers_count":1171,"open_issues_count":59,"forks_count":154,"subscribers_count":20,"default_branch":"main","last_synced_at":"2024-10-22T22:49:28.685Z","etag":null,"topics":["3d-modelling","3d-reconstruction","3d-rendering"],"latest_commit_sha":null,"homepage":"","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/apchenstu.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":"2022-03-17T18:21:32.000Z","updated_at":"2024-10-21T11:01:58.000Z","dependencies_parsed_at":"2024-01-18T18:07:59.547Z","dependency_job_id":"3cc45760-1339-4822-96c0-5b8460460f6a","html_url":"https://github.com/apchenstu/TensoRF","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/apchenstu%2FTensoRF","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/apchenstu%2FTensoRF/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/apchenstu%2FTensoRF/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/apchenstu%2FTensoRF/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/apchenstu","download_url":"https://codeload.github.com/apchenstu/TensoRF/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":221800051,"owners_count":16882469,"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":["3d-modelling","3d-reconstruction","3d-rendering"],"created_at":"2024-07-31T04:01:49.018Z","updated_at":"2024-10-28T07:30:23.823Z","avatar_url":"https://github.com/apchenstu.png","language":"Python","funding_links":[],"categories":["Papers"],"sub_categories":[],"readme":"# TensoRF\n## [Project page](https://apchenstu.github.io/TensoRF/) |  [Paper](https://arxiv.org/abs/2203.09517)\nThis repository contains a pytorch implementation for the paper: [TensoRF: Tensorial Radiance Fields](https://arxiv.org/abs/2203.09517). Our work present a novel approach to model and reconstruct radiance fields, which achieves super\n**fast** training process, **compact** memory footprint and **state-of-the-art** rendering quality.\u003cbr\u003e\u003cbr\u003e\n\n\nhttps://user-images.githubusercontent.com/16453770/158920837-3fafaa17-6ed9-4414-a0b1-a80dc9e10301.mp4\n## Installation\n\n#### Tested on Ubuntu 20.04 + Pytorch 1.10.1 \n\nInstall environment:\n```\nconda create -n TensoRF python=3.8\nconda activate TensoRF\npip install torch torchvision\npip install tqdm scikit-image opencv-python configargparse lpips imageio-ffmpeg kornia lpips tensorboard\n```\n\n\n## Dataset\n* [Synthetic-NeRF](https://drive.google.com/drive/folders/128yBriW1IG_3NJ5Rp7APSTZsJqdJdfc1) \n* [Synthetic-NSVF](https://dl.fbaipublicfiles.com/nsvf/dataset/Synthetic_NSVF.zip)\n* [Tanks\u0026Temples](https://dl.fbaipublicfiles.com/nsvf/dataset/TanksAndTemple.zip)\n* [Forward-facing](https://drive.google.com/drive/folders/128yBriW1IG_3NJ5Rp7APSTZsJqdJdfc1)\n\n\n\n## Quick Start\nThe training script is in `train.py`, to train a TensoRF:\n\n```\npython train.py --config configs/lego.txt\n```\n\n\nwe provide a few examples in the configuration folder, please note:\n\n `dataset_name`, choices = ['blender', 'llff', 'nsvf', 'tankstemple'];\n\n `shadingMode`, choices = ['MLP_Fea', 'SH'];\n\n `model_name`, choices = ['TensorVMSplit', 'TensorCP'], corresponding to the VM and CP decomposition. \n You need to uncomment the last a few rows of the configuration file if you want to training with the TensorCP model；\n\n `n_lamb_sigma` and `n_lamb_sh` are string type refer to the basis number of density and appearance along XYZ\ndimension;\n\n `N_voxel_init` and `N_voxel_final` control the resolution of matrix and vector;\n\n `N_vis` and `vis_every` control the visualization during training;\n\n  You need to set `--render_test 1`/`--render_path 1` if you want to render testing views or path after training. \n\nMore options refer to the `opt.py`. \n\n### For pretrained checkpoints and results please see:\n[https://1drv.ms/u/s!Ard0t_p4QWIMgQ2qSEAs7MUk8hVw?e=dc6hBm](https://1drv.ms/u/s!Ard0t_p4QWIMgQ2qSEAs7MUk8hVw?e=dc6hBm)\n\n\n\n## Rendering\n\n```\npython train.py --config configs/lego.txt --ckpt path/to/your/checkpoint --render_only 1 --render_test 1 \n```\n\nYou can just simply pass `--render_only 1` and `--ckpt path/to/your/checkpoint` to render images from a pre-trained\ncheckpoint. You may also need to specify what you want to render, like `--render_test 1`, `--render_train 1` or `--render_path 1`.\nThe rendering results are located in your checkpoint folder. \n\n## Extracting mesh\nYou can also export the mesh by passing `--export_mesh 1`:\n```\npython train.py --config configs/lego.txt --ckpt path/to/your/checkpoint --export_mesh 1\n```\nNote: Please re-train the model and don't use the pretrained checkpoints provided by us for mesh extraction, \nbecause some render parameters has changed.\n\n## Training with your own data\nWe provide two options for training on your own image set:\n\n1. Following the instructions in the [NSVF repo](https://github.com/facebookresearch/NSVF#prepare-your-own-dataset), then set the dataset_name to 'tankstemple'.\n2. Calibrating images with the script from [NGP](https://github.com/NVlabs/instant-ngp/blob/master/docs/nerf_dataset_tips.md):\n`python dataLoader/colmap2nerf.py --colmap_matcher exhaustive --run_colmap`, then adjust the datadir in `configs/your_own_data.txt`. Please check the `scene_bbox` and `near_far` if you get abnormal results.\n    \n\n## Citation\nIf you find our code or paper helps, please consider citing:\n```\n@INPROCEEDINGS{Chen2022ECCV,\n  author = {Anpei Chen and Zexiang Xu and Andreas Geiger and Jingyi Yu and Hao Su},\n  title = {TensoRF: Tensorial Radiance Fields},\n  booktitle = {European Conference on Computer Vision (ECCV)},\n  year = {2022}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/apchenstu.github.io%2FTensoRF%2F","html_url":"https://awesome.ecosyste.ms/projects/apchenstu.github.io%2FTensoRF%2F","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/apchenstu.github.io%2FTensoRF%2F/lists"}