{"id":20434197,"url":"https://github.com/vt-vl-lab/drg","last_synced_at":"2025-07-31T01:40:15.083Z","repository":{"id":54678376,"uuid":"289828346","full_name":"vt-vl-lab/DRG","owner":"vt-vl-lab","description":"[ECCV 2020] DRG: Dual Relation Graph for Human-Object Interaction Detection","archived":false,"fork":false,"pushed_at":"2022-03-11T16:50:53.000Z","size":1390,"stargazers_count":67,"open_issues_count":1,"forks_count":19,"subscribers_count":7,"default_branch":"master","last_synced_at":"2025-04-12T21:11:37.313Z","etag":null,"topics":["graph-network","human-object-interaction"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":false,"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/vt-vl-lab.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}},"created_at":"2020-08-24T04:33:19.000Z","updated_at":"2025-02-24T03:53:20.000Z","dependencies_parsed_at":"2022-08-13T23:40:54.078Z","dependency_job_id":null,"html_url":"https://github.com/vt-vl-lab/DRG","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/vt-vl-lab/DRG","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vt-vl-lab%2FDRG","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vt-vl-lab%2FDRG/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vt-vl-lab%2FDRG/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vt-vl-lab%2FDRG/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/vt-vl-lab","download_url":"https://codeload.github.com/vt-vl-lab/DRG/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vt-vl-lab%2FDRG/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":267973196,"owners_count":24174399,"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":"2025-07-30T02:00:09.044Z","response_time":70,"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":["graph-network","human-object-interaction"],"created_at":"2024-11-15T08:25:01.263Z","updated_at":"2025-07-31T01:40:15.058Z","avatar_url":"https://github.com/vt-vl-lab.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# DRG: Dual Relation Graph for Human-Object Interaction Detection\nOfficial Pytorch implementation for [DRG: Dual Relation Graph for Human-Object Interaction Detection (ECCV 2020)](https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123570681.pdf).\n\nSee the [project page](http://chengao.vision/DRG/) for more details. Please contact Jiarui Xu (jiaruixu@vt.edu) if you have any questions related to implementation details.\n\n\u003cimg src='demo/teaser.png'\u003e\n\n### Prerequisites\nThis codebase was tested with Python 3.6, Pytorch 1.0 from a nightly release, CUDA 10.0, and CentOS 7.4.1708.\n\n### Installation\nPlease check [INSTALL.md](INSTALL.md) for installation instructions.\n\n### Data Downloads\nDownload V-COCO and HICO-DET data. Setup HICO-DET evaluation code.\n```Shell\nbash ./scripts/download_dataset.sh \nbash ./scripts/download_data.sh\n```\n\n### Evaluation\n1. Download DRG detections and data\n    ```Shell\n    bash ./scripts/download_drg_detection.sh\n    ```\n\n2. Evaluate on VCOCO\n    ```Shell\n    python tools/vcoco_compute_mAP.py \\\n        --dataset_name vcoco_test \\\n        --detection_file output/VCOCO/detection_merged_human_object_app.pkl\n    ```\n\n3. Evaluate on HICO-DET\n    ```Shell\n    cd Data/ho-rcnn\n    matlab -r \"Generate_detection('COCO'); quit\"\n    cd ../../\n    ```\n\n4. Evaluate on HICO-DET finetuned detection\n    ```Shell\n    cd Data/ho-rcnn\n    matlab -r \"Generate_detection('finetune'); quit\"\n    cd ../../\n    ```\n\n### Train\n0. Down pre-trained Faster R-CNN model weights for initialization\n    ```Shell\n    bash ./scripts/download_frcnn.sh\n    ```\n\n1. Train on V-COCO\n    ```Shell\n    bash ./scripts/train_VCOCO.sh\n    ```\n\n2. Train on HICO-DET\n    ```Shell\n    bash ./scripts/train_HICO.sh\n    ```\n\n### Test\n1. Test on V-COCO\n    ```Shell\n    bash ./scripts/test_VCOCO.sh $APP_ITER_NUMBER $HUMAN_SP_ITER_NUMBER $OBJECT_SP_ITER_NUMBER\n    ```\n\n2. Test on HICO-DET\n    ```Shell\n    bash ./scripts/test_HICO.sh $APP_ITER_NUMBER $HUMAN_SP_ITER_NUMBER $OBJECT_SP_ITER_NUMBER\n    ```\n\n3. Test on HICO-DET w/ a fine-tined detector\n    ```Shell\n    bash ./scripts/test_HICO_ft.sh\n    ```\n\n**NOTE:** If you wish the use the same detector for a fair comparison, see [here](DETECTOR.md).\n\n\n### DRG Pretrained Weights\nDownload DRG trained weights.\n```Shell\nbash ./scripts/download_drg_models.sh\n```\n\n### Object Detection\nFor a simple demo, you can try\n```Shell\npython demo/demo_obj_det.py\n```\nCurrently, we only support Faster R-CNN with ResNet-R50-FPN backbone.\n\n\n### Citation\nIf you find this code useful for your research, please consider citing the following papers:\n\n\t@inproceedings{Gao-ECCV-DRG,\n\t    author    = {Gao, Chen and Xu, Jiarui and Zou, Yuliang and Huang, Jia-Bin}, \n\t    title     = {DRG: Dual Relation Graph for Human-Object Interaction Detection}, \n\t    booktitle = {European Conference on Computer Vision},\n\t    year      = {2020}\n\t}\n\n\t@inproceedings{gao2018ican,\n\t    author    = {Gao, Chen and Zou, Yuliang and Huang, Jia-Bin}, \n\t    title     = {iCAN: Instance-Centric Attention Network for Human-Object Interaction Detection}, \n\t    booktitle = {British Machine Vision Conference},\n\t    year      = {2018}\n\t}\n\n### Acknowledgement\nThis code follows the implementation architecture of [maskrcnn-benchmark](https://github.com/facebookresearch/maskrcnn-benchmark), [iCAN](https://github.com/vt-vl-lab/iCAN) and [No Frills](https://github.com/BigRedT/no_frills_hoi_det).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvt-vl-lab%2Fdrg","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvt-vl-lab%2Fdrg","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvt-vl-lab%2Fdrg/lists"}