{"id":13641436,"url":"https://github.com/misads/easy_detection","last_synced_at":"2025-12-25T08:10:05.378Z","repository":{"id":37630802,"uuid":"297974550","full_name":"misads/easy_detection","owner":"misads","description":"一个简单方便的目标检测框架(PyTorch环境可直接运行，不需要cuda编译)，支持Faster_RCNN、Cascade_RCNN、Yolo系列、SSD等经典网络。","archived":false,"fork":false,"pushed_at":"2024-09-19T02:54:41.000Z","size":7022,"stargazers_count":269,"open_issues_count":35,"forks_count":41,"subscribers_count":4,"default_branch":"master","last_synced_at":"2025-04-20T07:41:19.786Z","etag":null,"topics":["coco","detection","pytorch","voc","yolo"],"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/misads.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,"zenodo":null}},"created_at":"2020-09-23T13:13:15.000Z","updated_at":"2025-04-09T00:53:00.000Z","dependencies_parsed_at":"2025-04-23T03:15:30.818Z","dependency_job_id":null,"html_url":"https://github.com/misads/easy_detection","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/misads/easy_detection","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/misads%2Feasy_detection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/misads%2Feasy_detection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/misads%2Feasy_detection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/misads%2Feasy_detection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/misads","download_url":"https://codeload.github.com/misads/easy_detection/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/misads%2Feasy_detection/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28024312,"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":"2025-12-25T02:00:05.988Z","response_time":58,"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":["coco","detection","pytorch","voc","yolo"],"created_at":"2024-08-02T01:01:20.719Z","updated_at":"2025-12-25T08:10:05.341Z","avatar_url":"https://github.com/misads.png","language":"Python","funding_links":[],"categories":["Other Versions of YOLO"],"sub_categories":[],"readme":"# easy_detection\n\n　　COCO和VOC目标检测，基于pytorch，开箱即用，不需要CUDA编译。支持Faster_RCNN、Cascade_RCNN、Yolo系列、SSD。\n\n　　对比mmdetection，mmdetection功能很多，但是封装的层数也过多，对于初学者不是太友好。因此将经典的检测模型用简单的方式整理或重写了一下。如果遇到问题欢迎提issue或者与我联系。\n\n　　Faster RCNN实现细节可以参考我的Blog：[Faster RCNN实现细节](http://wiki.xyu.ink/#/cv/faster_rcnn?id=a31-faster-rcnn).\n\n## 目录\n- [介绍](#介绍)\n- [使用说明](#使用说明)\n- [模型指标和预训练模型](#模型指标和预训练模型)\n- [参考链接](#参考链接)\n\n## 介绍\n\n　　支持结果可视化、自定义数据集、多卡同步训练。  \n\n　　训练时间(Faster RCNN)：\n\n||单卡2080ti|8卡2080ti|\n|---|---|---|\n|VOC|6小时|45分钟|\n|COCO|48小时|6小时|\n\n## 使用说明\n\n- 安装和使用方法见 [使用手册.md](/_assets/_docs/tutorial.md).\n\n- Faster RCNN实现细节请参考 [Faster RCNN实现细节](http://wiki.xyu.ink/#/cv/faster_rcnn?id=a31-faster-rcnn).\n\n- 带有详细注释的代码细节请参考 `network/Faster_RCNN_v2/faster_rcnn`目录下的相关文件。\n\n## 模型指标和预训练模型\n\n### VOC数据集\n\n| 结构 | mAP@.5 | 下载链接 | 密码 | 日志 |\n| ----------- | -------- | ----- | ----- | ----- |\n| YoloV2  | 76.46|   [[百度网盘]](https://pan.baidu.com/s/1UyWGG1kn5h1l_FHP3idurw)| mwik | [[训练日志]](/_assets/_logs/yolo2_voc.txt) |\n| FasterRCNN + Res50 + FPN | 82.39 |  [[百度网盘]](https://pan.baidu.com/s/17NDNGeVRYxCG0vWqgaFDxQ) | isqt |  [[训练日志]](/_assets/_logs/faster_rcnn_voc.txt) |\n| CascadeRCNN + Res50 + FPN | 81.90 |  - | - | [[训练日志]](/_assets/_logs/cascade_rcnn_voc.txt)  |\n| SSD300 + VGG16 | 79.21 | [[百度网盘]](https://pan.baidu.com/s/18XN0Atybz27DnwFdUsMRPg)| 59y0 | - |\n| SSD512 + VGG16 |   82.14 | [[百度网盘]](https://pan.baidu.com/s/1CYB7GvLYxin01Oqwo0v7ZQ)| 0iur | - |\n\n\n\n### COCO数据集\n\n\n| 结构 | COCO AP\\* | mAP@.5 | mAP@.75 |下载链接 | 日志 |\n| --------------- | ---------- | ------ | ----- | ----- | ----- |\n| FasterRCNN + Res50 + FPN | 35.41 |57.11| 38.43 | [[pytorch]](https://download.pytorch.org/models/fasterrcnn_resnet50_fpn_coco-258fb6c6.pth) | [[训练日志]](/_assets/_logs/faster_rcnn_coco.txt) |\n| CascadeRCNN + Res50 + FPN | 38.71 |56.61| 42.16 | - | [[训练日志]](/_assets/_logs/cascade_rcnn_coco.txt) |\n\n\n　　\\*注：COCO AP是IoU@\\[0.5:0.95\\]的mAP平均值。\n\n## 参考链接\n\n- SSD \u003chttps://github.com/lufficc/SSD\u003e\n  \n- YoloV2、YoloV3 \u003chttps://github.com/andy-yun/pytorch-0.4-yolov3\u003e\n\n- EfficientDet \u003chttps://github.com/rwightman/efficientdet-pytorch\u003e\n\n- YoloV4 \u003chttps://github.com/Tianxiaomo/pytorch-YOLOv4\u003e \u003chttps://github.com/argusswift/YOLOv4-pytorch\u003e\n\n- YoloV5 \u003chttps://github.com/ultralytics/yolov5\u003e\n\n- Faster_RCNN \u003chttps://github.com/pytorch/vision/tree/master/torchvision/models/detection\u003e\n\n- RetinaNet \u003chttps://github.com/yhenon/pytorch-retinanet\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmisads%2Feasy_detection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmisads%2Feasy_detection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmisads%2Feasy_detection/lists"}