{"id":13445345,"url":"https://github.com/Arthur151/ROMP","last_synced_at":"2025-03-20T20:32:25.553Z","repository":{"id":37440246,"uuid":"289262638","full_name":"Arthur151/ROMP","owner":"Arthur151","description":"Monocular, One-stage, Regression of Multiple 3D People and their 3D positions \u0026 trajectories in camera \u0026 global coordinates. ROMP[ICCV21], BEV[CVPR22], TRACE[CVPR2023]","archived":false,"fork":false,"pushed_at":"2023-10-26T15:45:31.000Z","size":85086,"stargazers_count":1344,"open_issues_count":188,"forks_count":229,"subscribers_count":41,"default_branch":"master","last_synced_at":"2024-10-16T14:06:09.374Z","etag":null,"topics":["3d-mesh-recovery","bottom-up","multi-person","multi-person-3d-mesh-recovery","pose-estimation","pytorch","smpl"],"latest_commit_sha":null,"homepage":"https://www.yusun.work/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Arthur151.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","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":"2020-08-21T12:18:59.000Z","updated_at":"2024-10-14T23:38:21.000Z","dependencies_parsed_at":"2024-01-07T04:49:18.922Z","dependency_job_id":null,"html_url":"https://github.com/Arthur151/ROMP","commit_stats":{"total_commits":312,"total_committers":10,"mean_commits":31.2,"dds":0.5512820512820513,"last_synced_commit":"12a624f00be92d1f7195f24106d151d2f26961f4"},"previous_names":[],"tags_count":8,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Arthur151%2FROMP","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Arthur151%2FROMP/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Arthur151%2FROMP/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Arthur151%2FROMP/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Arthur151","download_url":"https://codeload.github.com/Arthur151/ROMP/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":221807657,"owners_count":16883629,"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-mesh-recovery","bottom-up","multi-person","multi-person-3d-mesh-recovery","pose-estimation","pytorch","smpl"],"created_at":"2024-07-31T05:00:31.450Z","updated_at":"2024-10-28T08:30:37.611Z","avatar_url":"https://github.com/Arthur151.png","language":"Python","funding_links":[],"categories":["六、PV2BEV method summary","Multi-Person 3D Mesh Recovery","Human Mesh Recovery"],"sub_categories":["1. Summary of PV2BEV based on depth method","2021","For Multiple Persons"],"readme":"| \u003ch2 align=\"center\"\u003e ROMP \u003c/h2\u003e | \u003ch2 align=\"center\"\u003e BEV \u003c/h2\u003e | \u003ch2 align=\"center\"\u003e TRACE \u003c/h2\u003e |\n| :---: | :---: | :---: |\n| Monocular, One-stage, Regression of Multiple 3D People (ICCV21) | Putting People in their Place: Monocular Regression of 3D People in Depth (CVPR2022) | TRACE: 5D Temporal Regression of Avatars with Dynamic Cameras in 3D Environments (CVPR2023) |\n| ROMP is a **one-stage** method for monocular multi-person 3D mesh recovery in **real time**. | BEV further explores multi-person **depth relationships** and supports **all age groups**. | TRACE further **tracks specific subjects** and recover their **global 3D trajectory with dynamic cameras**. |\n| **[[Paper]](https://arxiv.org/abs/2008.12272) [[Video]](https://www.youtube.com/watch?v=hunBPJxnyBU)** | **[[Project Page]](https://arthur151.github.io/BEV/BEV.html) [[Paper]](https://arxiv.org/abs/2112.08274) [[Video]](https://youtu.be/Q62fj_6AxRI)** |  **[[Project Page]](https://arthur151.github.io/TRACE/TRACE.html) [[Paper]](http://arxiv.org/abs/2306.02850) [[Video]](https://www.youtube.com/watch?v=l8aLHDXWQRw)** |\n| | **[[RelativeHuman Dataset]](https://github.com/Arthur151/Relative_Human)** | **[[DynaCam Dataset]](https://github.com/Arthur151/DynaCam)** |\n| \u003cimg src=\"../assets/demo/animation/blender_character_driven-min.gif\" alt=\"drawing\" height=\"230\"/\u003e | \u003cimg src=\"../assets/demo/images_results/BEV_tennis_results.png\" alt=\"drawing\" height=\"230\"/\u003e | \u003cimg src=\"https://www.yusun.work/TRACE/images/demo.gif\" alt=\"drawing\" height=\"230\"/\u003e |\n\nWe provide **cross-platform API** (installed via pip) to run ROMP \u0026 BEV on Linux / Windows / Mac. \n\n## Table of contents\n- [Table of contents](#table-of-contents)\n- [News](#news)\n- [Getting started](#getting-started)\n  - [Installation](#installation)\n  - [Try on Google Colab](#try-on-google-colab)\n- [How to use it](#how-to-use-it)\n    - [Please refer to this guidance for inference \u0026 export (fbx/glb/bvh).](#please-refer-to-this-guidance-for-inference--export-fbxglbbvh)\n  - [Train](#train)\n  - [Evaluation](#evaluation)\n  - [Docker usage](#docker-usage)\n  - [Bugs report](#bugs-report)\n- [Citation](#citation)\n- [Acknowledgement](#acknowledgement)\n\n## News\n*2023/06/17: Release of TRACE's code. Please refer to this [instructions](simple_romp/trace2/README.md) for inference.*   \n*2022/06/21: Training \u0026 evaluation code of BEV is released. Please update the [model_data](https://github.com/Arthur151/ROMP/releases/download/v1.1/model_data.zip).*   \n*2022/05/16: simple-romp v1.0 is released to support tracking, calling in python, exporting bvh, and etc.*   \n*2022/04/14: Inference code of BEV has been released in simple-romp v0.1.0.*   \n*2022/04/10: Adding onnx support, with faster inference speed on CPU/GPU.*   \n[Old logs](docs/updates.md)\n\n## Getting started\n\nPlease use simple-romp for inference, the rest code is just for training.\n\n## How to use it\n\n## ROMP \u0026 BEV\n#### For inference \u0026 export (fbx/glb/bvh), please refer to [this guidance](https://github.com/Arthur151/ROMP/blob/master/simple_romp/README.md).\n#### For training, please refer to [installation.md](docs/installation.md) for full installation, [dataset.md](docs/dataset.md) for data preparation, [train.md](docs/train.md) for training.\n#### For evaluation on benchmarks, please refer to [romp_evaluation](docs/romp_evaluation.md), [bev_evaluation](docs/bev_evaluation.md).\n\n## TRACE\n#### For inference, please refer to [this instrcution](simple_romp/trace2/README.md).\n#### For evaluation on benchmarks, please refer to [trace_evaluation](simple_romp/trace2/README.md).\n#### For training, please refer to [trace_train](trace/README.md).\n\n### Extensions\n\n[[Blender addon]](https://github.com/yanchxx/CDBA): [Yan Chuanhang](https://github.com/yanchxx) created a Blender-addon to drive a 3D character in Blender using ROMP from image, video or webcam input.\n\n[[VMC protocol]](https://codeberg.org/vivi90/vmcps): [Vivien Richter](https://github.com/vivi90) implemented a VMC (Virtual Motion Capture) protocol support for different Motion Capture solutions with ROMP. \n\n### Docker usage\n\nPlease refer to [docker.md](docs/docker.md)\n\n### Bugs report\n\nWelcome to submit issues for the bugs.\n\n## Contributors\n\nThis repository is maintained by [Yu Sun](https://www.yusun.work/).  \n\nROMP has also benefited from many developers, including   \n - [Peng Cheng](https://github.com/CPFLAME) : constructive discussion on Center map training.  \n - [Marco Musy](https://github.com/marcomusy) : help in [the textured SMPL visualization](https://github.com/marcomusy/vedo/issues/371).  \n - [Gavin Gray](https://github.com/gngdb) : adding support for an elegant context manager to run code in a notebook.  \n - [VLT Media](https://github.com/vltmedia) and [Vivien Richter](https://github.com/vivi90) : adding support for running on Windows \u0026 batch_videos.py.  \n - [Chuanhang Yan](https://github.com/yanch2116) : developing an [addon for driving character in Blender](https://github.com/yanch2116/Blender-addons-for-SMPL).  \n - [Tian Jin](https://github.com/jinfagang): help in simplified smpl and fast rendering ([realrender](https://pypi.org/project/realrender/)).\n - [ZhengdiYu](https://github.com/ZhengdiYu) : helpful discussion on optimizing the implementation details.\n - [Ali Yaghoubian](https://github.com/AliYqb) : add Docker file for simple-romp.\n\n## Citation\n```bibtex\n@InProceedings{TRACE,\n    author = {Sun, Yu and Bao, Qian and Liu, Wu and Mei, Tao and Black, Michael J.},\n    title = {{TRACE: 5D Temporal Regression of Avatars with Dynamic Cameras in 3D Environments}}, \n    booktitle = {CVPR}, \n    year = {2023}}\n@InProceedings{BEV,\n    author = {Sun, Yu and Liu, Wu and Bao, Qian and Fu, Yili and Mei, Tao and Black, Michael J},\n    title = {{Putting People in their Place: Monocular Regression of 3D People in Depth}},\n    booktitle = {CVPR},\n    year = {2022}}\n@InProceedings{ROMP,\n    author = {Sun, Yu and Bao, Qian and Liu, Wu and Fu, Yili and Michael J., Black and Mei, Tao},\n    title = {{Monocular, One-stage, Regression of Multiple 3D People}},\n    booktitle = {ICCV},\n    year = {2021}}\n```\n\n## Acknowledgement\nThis work was supported by the National Key R\u0026D Program of China under Grand No. 2020AAA0103800.  \n**MJB Disclosure**: [https://files.is.tue.mpg.de/black/CoI_CVPR_2023.txt](https://files.is.tue.mpg.de/black/CoI_CVPR_2023.txt)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FArthur151%2FROMP","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FArthur151%2FROMP","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FArthur151%2FROMP/lists"}