{"id":50612111,"url":"https://github.com/THU-DA-6D-Pose-Group/GDR-Net","last_synced_at":"2026-06-23T00:00:54.416Z","repository":{"id":38365502,"uuid":"340921779","full_name":"THU-DA-6D-Pose-Group/GDR-Net","owner":"THU-DA-6D-Pose-Group","description":"GDR-Net: Geometry-Guided Direct Regression Network for Monocular 6D Object Pose Estimation. (CVPR 2021) ","archived":false,"fork":false,"pushed_at":"2023-11-28T05:38:02.000Z","size":1009,"stargazers_count":214,"open_issues_count":4,"forks_count":39,"subscribers_count":11,"default_branch":"main","last_synced_at":"2023-11-28T06:30:34.743Z","etag":null,"topics":["6d-pose-estimation","detectron2","mmcv"],"latest_commit_sha":null,"homepage":"https://github.com/THU-DA-6D-Pose-Group/GDR-Net","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/THU-DA-6D-Pose-Group.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}},"created_at":"2021-02-21T14:31:57.000Z","updated_at":"2023-11-24T20:11:49.000Z","dependencies_parsed_at":"2023-11-28T06:39:53.389Z","dependency_job_id":null,"html_url":"https://github.com/THU-DA-6D-Pose-Group/GDR-Net","commit_stats":null,"previous_names":[],"tags_count":3,"template":null,"template_full_name":null,"purl":"pkg:github/THU-DA-6D-Pose-Group/GDR-Net","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/THU-DA-6D-Pose-Group%2FGDR-Net","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/THU-DA-6D-Pose-Group%2FGDR-Net/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/THU-DA-6D-Pose-Group%2FGDR-Net/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/THU-DA-6D-Pose-Group%2FGDR-Net/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/THU-DA-6D-Pose-Group","download_url":"https://codeload.github.com/THU-DA-6D-Pose-Group/GDR-Net/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/THU-DA-6D-Pose-Group%2FGDR-Net/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34669839,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-22T02:00:06.391Z","response_time":106,"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":["6d-pose-estimation","detectron2","mmcv"],"created_at":"2026-06-06T05:00:21.752Z","updated_at":"2026-06-23T00:00:54.411Z","avatar_url":"https://github.com/THU-DA-6D-Pose-Group.png","language":"Python","funding_links":[],"categories":["6D Object Pose Estimation"],"sub_categories":["Methods"],"readme":"# GDR-Net\nThis repo provides the PyTorch implementation of the work:\n\n**Gu Wang, Fabian Manhardt, Federico Tombari, Xiangyang Ji. GDR-Net: Geometry-Guided Direct Regression Network for Monocular 6D Object Pose Estimation. In CVPR 2021.**\n[[Paper](https://openaccess.thecvf.com/content/CVPR2021/html/Wang_GDR-Net_Geometry-Guided_Direct_Regression_Network_for_Monocular_6D_Object_Pose_CVPR_2021_paper.html)][[ArXiv](http://arxiv.org/abs/2102.12145)][[Video](https://www.bilibili.com/video/BV1dU4y1G7Ku?share_source=copy_web)][[bibtex](#Citation)]\n\n\n## News\n* [2023/10] An enhanced version of this work, [GPose2023](https://zhangcyg.github.io/figs/GPose_ICCV23.pdf) by Zhang et al., won BOP Challenge @ ICCV 2023. Congratulations!\n* [2022/10] An enhanced version of this work, [GDRNPP](https://github.com/shanice-l/gdrnpp_bop2022.git) by Liu et al., won most of the [awards](http://cmp.felk.cvut.cz/sixd/workshop_2022/slides/bop_challenge_2022_results.pdf) on [BOP Challenge @ ECCV 2022](https://bop.felk.cvut.cz/challenges/bop-challenge-2022/). Congratulations!\n* [2021/08] An extension of this work, [SO-Pose](https://arxiv.org/abs/2108.08367) by Di et al. (ICCV 2021), has been released ([SO-Pose code](https://github.com/shangbuhuan13/SO-Pose), [mirror](https://github.com/THU-DA-6D-Pose-Group/SO-Pose)).\n\n## Overview\n\u003cp align=\"center\"\u003e\n\u003cimg src='assets/gdrn_architecture.png' width='800'\u003e\n\u003cp\u003e\n\n\n\n## Requirements\n* Ubuntu 16.04/18.04, CUDA 10.1/10.2, python \u003e= 3.6, PyTorch \u003e= 1.6, torchvision\n* Install `detectron2` from [source](https://github.com/facebookresearch/detectron2)\n* `sh scripts/install_deps.sh`\n* Compile the cpp extension for `farthest points sampling (fps)`:\n    ```\n    sh core/csrc/compile.sh\n    ```\n\n## Datasets\nDownload the 6D pose datasets (LM, LM-O, YCB-V) from the\n[BOP website](https://bop.felk.cvut.cz/datasets/) and\n[VOC 2012](https://pjreddie.com/projects/pascal-voc-dataset-mirror/)\nfor background images.\nPlease also download the `image_sets` and `test_bboxes` from\nhere ([BaiduNetDisk](https://pan.baidu.com/s/1gGoZGkuMYxhU9LBKxuSz0g), password: qjfk | [Cloud.THU](https://cloud.tsinghua.edu.cn/d/b2311297acf54f26b429/), password: fMNOASFHW0E8R72357T6mn9).\n\nThe structure of `datasets` folder should look like below:\n```\n# recommend using soft links (ln -sf)\ndatasets/\n├── BOP_DATASETS\n    ├──lm\n    ├──lmo\n    ├──ycbv\n├── lm_imgn  # the OpenGL rendered images for LM, 1k/obj\n├── lm_renders_blender  # the Blender rendered images for LM, 10k/obj (pvnet-rendering)\n├── VOCdevkit\n```\n\n* `lm_imgn` comes from [DeepIM](https://github.com/liyi14/mx-DeepIM), which can be downloaded here ([BaiduNetDisk](https://pan.baidu.com/s/1e9SJoqb0EmyqVLEVlbNQIA), password: vr0i | [Cloud.THU](https://cloud.tsinghua.edu.cn/f/22c4fba9c06f47d3aba4/), password: A097fN8a07ufn0u70).\n\n* `lm_renders_blender` comes from [pvnet-rendering](https://github.com/zju3dv/pvnet-rendering), note that we do not need the fused data.\n\n\n## Training GDR-Net\n`./core/gdrn_modeling/train_gdrn.sh \u003cconfig_path\u003e \u003cgpu_ids\u003e (other args)`\n\nExample:\n```\n./core/gdrn_modeling/train_gdrn.sh configs/gdrn/lm/a6_cPnP_lm13.py 0  # multiple gpus: 0,1,2,3\n# add --resume if you want to resume from an interrupted experiment.\n```\n\n\nOur trained GDR-Net models can be found here ([BaiduNetDisk](https://pan.baidu.com/s/1_MEZJBd67hdxcE8JzmnOtA), password: kedv | [Cloud.THU](https://cloud.tsinghua.edu.cn/d/b17100d09a9f4a208549/), password: 0M,oadfu9uIU). \u003cbr /\u003e\n\u003csub\u003e\u003csup\u003e(Note that the models for BOP setup in the supplement were trained using a refactored version of this repo (not compatible), they are slightly better than the models provided here.)\u003c/sup\u003e\u003c/sub\u003e\n\n\n## Evaluation\n`./core/gdrn_modeling/test_gdrn.sh \u003cconfig_path\u003e \u003cgpu_ids\u003e \u003cckpt_path\u003e (other args)`\n\nExample:\n```\n./core/gdrn_modeling/test_gdrn.sh configs/gdrn/lmo/a6_cPnP_AugAAETrunc_BG0.5_lmo_real_pbr0.1_40e.py 0 output/gdrn/lmo/a6_cPnP_AugAAETrunc_BG0.5_lmo_real_pbr0.1_40e/gdrn_lmo_real_pbr.pth\n```\n\n\n## Citation\nIf you find this useful in your research, please consider citing:\n```\n@InProceedings{Wang_2021_GDRN,\n    title     = {{GDR-Net}: Geometry-Guided Direct Regression Network for Monocular 6D Object Pose Estimation},\n    author    = {Wang, Gu and Manhardt, Fabian and Tombari, Federico and Ji, Xiangyang},\n    booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},\n    month     = {June},\n    year      = {2021},\n    pages     = {16611-16621}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FTHU-DA-6D-Pose-Group%2FGDR-Net","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FTHU-DA-6D-Pose-Group%2FGDR-Net","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FTHU-DA-6D-Pose-Group%2FGDR-Net/lists"}