{"id":14260811,"url":"https://github.com/xingy038/ctnerf","last_synced_at":"2025-08-13T09:31:50.810Z","repository":{"id":246416980,"uuid":"800012735","full_name":"xingy038/CTNeRF","owner":"xingy038","description":null,"archived":false,"fork":false,"pushed_at":"2024-07-16T19:00:13.000Z","size":92,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2024-08-22T13:14:54.334Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/xingy038.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2024-05-13T14:32:59.000Z","updated_at":"2024-07-28T12:25:50.000Z","dependencies_parsed_at":"2024-06-27T22:01:38.086Z","dependency_job_id":"92187b20-4962-4048-8fc9-e1fc46dc86cb","html_url":"https://github.com/xingy038/CTNeRF","commit_stats":null,"previous_names":["xingy038/ctnerf"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xingy038%2FCTNeRF","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xingy038%2FCTNeRF/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xingy038%2FCTNeRF/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xingy038%2FCTNeRF/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/xingy038","download_url":"https://codeload.github.com/xingy038/CTNeRF/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":229754429,"owners_count":18119124,"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":"2024-08-22T13:00:33.887Z","updated_at":"2024-12-14T20:30:30.845Z","avatar_url":"https://github.com/xingy038.png","language":"Python","funding_links":[],"categories":["2024"],"sub_categories":["Real"],"readme":"# CTNeRF: Cross-Time Transformer for Dynamic Neural Radiance Field from Monocular Video\n\n[![arXiv](https://img.shields.io/badge/arXiv-2401.04861-b31b1b.svg)](https://arxiv.org/abs/2401.04861)\n\n\u003e **CTNeRF: Cross-Time Transformer for Dynamic Neural Radiance Field from Monocular Video**\u003cbr\u003e\n\u003e [Paper(Arxiv)](https://arxiv.org/abs/2401.04861) | [Results](https://drive.google.com/file/d/10LHsemH6ImE4mghYImPtsVT1EpnbaNk5/view?usp=sharing)\u003cbr\u003e\n\u003e Xingyu Miao, Yang Bai, Haoran Duan, Yawen Huang, Fan Wan, Yang Long, Yefeng Zheng\u003cbr\u003e\n\u003e Accepted by Pattern Recognition (PR)\n\n\n\n## Setup\nThe code is test with\n* Linux \n* Anaconda 3\n* Python 3.8\n* CUDA 11.8\n* 2 3090 GPU\n\n\nTo get started, please create the conda environment `ctnerf` by running\n```\nconda create --name ctnerf python=3.8\nconda activate ctnerf\nconda install pytorch=1.13.1 torchvision=0.14.1 cudatoolkit=11.8 matplotlib tensorboard scipy opencv -c pytorch\npip install imageio scikit-image configargparse timm lpips\n```\nand install [COLMAP](https://colmap.github.io/install.html) manually. Then download MiDaS and RAFT weights\n```\nROOT_PATH=/path/to/the/CTNeRF/folder\ncd $ROOT_PATH\nwget --no-check-certificate https://filebox.ece.vt.edu/~chengao/free-view-video/weights.zip\nunzip weights.zip\nrm weights.zip\n```\n\n## Dynamic Scene Dataset\nThe [Dynamic Scene Dataset](https://www-users.cse.umn.edu/~jsyoon/dynamic_synth/) is used to\nquantitatively evaluate our method. Please download the pre-processed data by running:\n```\ncd $ROOT_PATH\nwget --no-check-certificate https://filebox.ece.vt.edu/~chengao/free-view-video/data.zip\nunzip data.zip\nrm data.zip\n```\n\n### Training\nYou can train a model from scratch by running:\n```\ncd $ROOT_PATH/\npython run_nerf.py --config configs/config_Balloon2.txt\n```\n\n### Rendering from pre-trained models\nWe also provide pre-trained models. You can download them by running:\n```\ncd $ROOT_PATH/\nwget --no-check-certificate https://filebox.ece.vt.edu/~chengao/free-view-video/logs.zip\nunzip logs.zip\nrm logs.zip\n```\n\nThen you can render the results directly by running:\n```\npython run_nerf.py --config configs/config_Balloon2.txt --render_only --ft_path $ROOT_PATH/logs/Balloon2_H270_CTNeRF_pretrain/300000.tar\n```\n\n### Evaluating\n\nPlease download the results by running:\n```\ncd $ROOT_PATH/\nwget --no-check-certificate https://filebox.ece.vt.edu/~chengao/free-view-video/results.zip\nunzip results.zip\nrm results.zip\n```\n\nThen you can calculate the PSNR/SSIM/LPIPS by running:\n```\ncd $ROOT_PATH/utils\npython evaluation.py\n```\n\n\nIf you find this code useful for your research, please consider citing the following paper:\n```\n@article{MIAO2024110729,\ntitle = {CTNeRF: Cross-time Transformer for dynamic neural radiance field from monocular video},\njournal = {Pattern Recognition},\npages = {110729},\nyear = {2024},\nissn = {0031-3203},\ndoi = {https://doi.org/10.1016/j.patcog.2024.110729},\nurl = {https://www.sciencedirect.com/science/article/pii/S0031320324004801},\nauthor = {Xingyu Miao and Yang Bai and Haoran Duan and Fan Wan and Yawen Huang and Yang Long and Yefeng Zheng},\nkeywords = {Dynamic neural radiance field, Monocular video, Scene flow, Transformer}\n}\n```\n## Acknowledgments\nOur training code is build upon\n[pixelNeRF](https://github.com/sxyu/pixel-nerf),\n[DynamicNeRF](https://github.com/gaochen315/DynamicNeRF), and\n[NSFF](https://github.com/zl548/Neural-Scene-Flow-Fields).\nOur flow prediction code is modified from [RAFT](https://github.com/princeton-vl/RAFT).\nOur depth prediction code is modified from [MiDaS](https://github.com/isl-org/MiDaS).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fxingy038%2Fctnerf","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fxingy038%2Fctnerf","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fxingy038%2Fctnerf/lists"}