{"id":20173509,"url":"https://github.com/moygcc/hpcwild","last_synced_at":"2025-09-29T08:24:38.488Z","repository":{"id":37728362,"uuid":"449694235","full_name":"MoyGcc/hpcwild","owner":"MoyGcc","description":"Human Performance Capture from Monocular Video in the Wild (3DV2021)","archived":false,"fork":false,"pushed_at":"2023-09-27T15:11:09.000Z","size":20679,"stargazers_count":85,"open_issues_count":0,"forks_count":9,"subscribers_count":7,"default_branch":"main","last_synced_at":"2025-04-10T03:37:35.228Z","etag":null,"topics":["3d-vision","3dv2021","computer-vision"],"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/MoyGcc.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":"2022-01-19T13:03:51.000Z","updated_at":"2024-12-13T00:51:31.000Z","dependencies_parsed_at":"2025-04-10T03:28:43.314Z","dependency_job_id":null,"html_url":"https://github.com/MoyGcc/hpcwild","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/MoyGcc/hpcwild","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MoyGcc%2Fhpcwild","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MoyGcc%2Fhpcwild/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MoyGcc%2Fhpcwild/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MoyGcc%2Fhpcwild/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/MoyGcc","download_url":"https://codeload.github.com/MoyGcc/hpcwild/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/MoyGcc%2Fhpcwild/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":277485339,"owners_count":25826051,"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","status":"online","status_checked_at":"2025-09-29T02:00:09.175Z","response_time":84,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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-vision","3dv2021","computer-vision"],"created_at":"2024-11-14T01:36:26.140Z","updated_at":"2025-09-29T08:24:38.467Z","avatar_url":"https://github.com/MoyGcc.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Human Performance Capture from Monocular Video in the Wild\n## [Paper](https://arxiv.org/pdf/2111.14672.pdf) | [Video](https://www.youtube.com/watch?v=5M7Ytnxmhd4) | [Project Page](https://ait.ethz.ch/human-performance-capture)\n\n\u003cimg src=\"assets/teaser.gif\" width=\"1000\" height=\"350\"/\u003e \n\nOfficial code release for 3DV 2021 paper [*Human Performance Capture from Monocular Video in the Wild*](https://arxiv.org/pdf/2111.14672.pdf). We propose a method capable of capturing the dynamic 3D human shape from a monocular video featuring challenging body poses, without any additional input.\n\nIf you find our code or paper useful, please cite as\n```\n@inproceedings{guo2021human,\n  title={Human Performance Capture from Monocular Video in the Wild},\n  author={Guo, Chen and Chen, Xu and Song, Jie and Hilliges, Otmar},\n  booktitle={2021 International Conference on 3D Vision (3DV)},\n  pages={889--898},\n  year={2021},\n  organization={IEEE}\n}\n```\n\n## Quick Start\nCLone this repo:\n```\ngit clone https://github.com/MoyGcc/hpcwild.git\ncd  hpcwild\nconda env create -f environment.yml\nconda activate hpcwild\n```\nAdditional Dependencies:\n1. Kaolin 0.1.0 (https://github.com/NVIDIAGameWorks/kaolin)\n2. MPI mesh library (https://github.com/MPI-IS/mesh)\n3. torch-mesh-isect (https://github.com/vchoutas/torch-mesh-isect)\n\nDownload [SMPL models](https://smpl.is.tue.mpg.de/downloads) (1.0.0 for Python 2.7 (10 shape PCs)) and move them to the corresponding places:\n```\nmkdir lib/smpl/smpl_model/\nmv /path/to/smpl/models/basicModel_f_lbs_10_207_0_v1.0.0.pkl smpl_rendering/smpl_model/SMPL_FEMALE.pkl\nmv /path/to/smpl/models/basicmodel_m_lbs_10_207_0_v1.0.0.pkl smpl_rendering/smpl_model/SMPL_MALE.pkl\n```\n\nDownload checkpoints for external modules:\n```\nwget https://download.01.org/opencv/openvino_training_extensions/models/human_pose_estimation/checkpoint_iter_370000.pth\nmv /path/to/checkpoint_iter_370000.pth external/lightweight-human-pose-estimation.pytorch/checkpoint_iter_370000.pth\n\nwget https://dl.fbaipublicfiles.com/pifuhd/checkpoints/pifuhd.pt pifuhd.pt \nmv /path/to/pifuhd.pt external/pifuhd/checkpoints/pifuhd.pt\n\nDownload IPNet weights: https://datasets.d2.mpi-inf.mpg.de/IPNet2020/IPNet_p5000_01_exp_id01.zip\nunzip IPNet_p5000_01_exp_id01.zip\nmv /path/to/IPNet_p5000_01_exp_id01 registration/experiments/IPNet_p5000_01_exp_id01\n\ngdown --id 1mcr7ALciuAsHCpLnrtG_eop5-EYhbCmz -O modnet_photographic_portrait_matting.ckpt\nmv /path/to/modnet_photographic_portrait_matting.ckpt external/MODNet/pretrained/modnet_photographic_portrait_matting.ckpt\n```\n### Test on 3DPW dataset\nDownload [3DPW dataset](https://virtualhumans.mpi-inf.mpg.de/3DPW/) \n1. modify the `dataset_path` in `test.conf`.\n2. run `bash mesh_recon.sh` to obtain the rigid body shape.\n3. run `bash registration.sh` to register a SMPL+D model to the rigid human body.\n4. run `bash tracking.sh` to capture the human performance temporally.\n\n### Test on your own video\n1. run [OpenPose](https://github.com/CMU-Perceptual-Computing-Lab/openpose) to obtain the 2D keypoints.\n2. run [LGD](https://github.com/InpatientJam/Learned-Gradient-Descent) to acquire the initial 3D poses.\n3. run [MODNet](https://github.com/ZHKKKe/MODNet) to extract sihouettes.\n\n# Acknowledgement\nWe use the code in [PIFuHD](https://github.com/facebookresearch/pifuhd) for the rigid body construction and adapt [IPNet](https://github.com/bharat-b7/IPNet) for human model registration. We use off-the-shelf methods [OpenPose](https://github.com/CMU-Perceptual-Computing-Lab/openpose) and [MODNet](https://github.com/ZHKKKe/MODNet) for the extraction of 2D keypoints and sihouettes. We sincerely thank these authors for their awesome work.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmoygcc%2Fhpcwild","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmoygcc%2Fhpcwild","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmoygcc%2Fhpcwild/lists"}