{"id":19320158,"url":"https://github.com/52cv/objectdetection","last_synced_at":"2026-02-15T16:08:03.822Z","repository":{"id":111173531,"uuid":"315247881","full_name":"52CV/ObjectDetection","owner":"52CV","description":null,"archived":false,"fork":false,"pushed_at":"2021-03-23T03:24:55.000Z","size":18,"stargazers_count":10,"open_issues_count":0,"forks_count":3,"subscribers_count":3,"default_branch":"main","last_synced_at":"2025-10-05T08:56:14.636Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/52CV.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2020-11-23T08:29:19.000Z","updated_at":"2022-01-20T18:59:15.000Z","dependencies_parsed_at":"2023-05-06T22:46:14.289Z","dependency_job_id":null,"html_url":"https://github.com/52CV/ObjectDetection","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/52CV/ObjectDetection","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/52CV%2FObjectDetection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/52CV%2FObjectDetection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/52CV%2FObjectDetection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/52CV%2FObjectDetection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/52CV","download_url":"https://codeload.github.com/52CV/ObjectDetection/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/52CV%2FObjectDetection/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29483518,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-15T15:33:17.885Z","status":"ssl_error","status_checked_at":"2026-02-15T15:32:53.698Z","response_time":118,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":[],"created_at":"2024-11-10T01:27:20.304Z","updated_at":"2026-02-15T16:08:03.793Z","avatar_url":"https://github.com/52CV.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Object Detection\n\n汇集目标检测顶端算法\u003cbr\u003e\n\n## 目录\n\n|:cat:|:dog:|:wolf:|\n|-----|-----|-----|\n|:free:|[EfficientDet](#5)|[YOLO 系列](#4)|\n|[NanoDet](#3)|[U^2-Net](#2)|[BASNet](#1)|\n\n\n\n\n\u003ca name=\"5\"/\u003e\n\n## 3.EfficientDet\n\n* 📜 [EfficientDet: Scalable and Efficient Object Detection](https://arxiv.org/abs/1911.09070)\n\u003cbr\u003eEfficientDet 是谷歌大脑团队在2020年3月发布的一个目标检测模型。与之前的检测器如 Mask R-CNN 相比，它以更少的参数和 FLOPs 达到了最先进的 53.7%COCO 平均精度（AP）。有 8 种变体，从 D0 到 D7，尺寸和精度都在不断改进。此外，最近发布了一个超大版本 D7x，达到 55.1% 的 AP。\u003cbr\u003e\n  * CVPR 2020\n  * :star:[Github](https://github.com/google/automl/tree/master/efficientdet)\n\n\u003ca name=\"4\"/\u003e\n\n## 2.YOLO 系列\n\n* 解读：[YOLO 系目标检测算法家族全景图！](https://mp.weixin.qq.com/s/wjUiSRcP49gkn5YwpPyPpA)\n\n* 📜 [Scaled-YOLOv4: Scaling Cross Stage Partial Network](https://arxiv.org/2011.08036)\u003cbr\u003e\n   * YOLOv4-large 在COCO数据集达到SOTA精度: 55.4% AP(73.3% AP50) 并以以15 fps 在 Tesla V100运行, 而如果加上测试阶段数据增强方法后,YOLOv4-large 达到 55.8% AP (73.2 AP50). 作者称这一精度是所有已公开文献的最高精度.\n   * 解读：[上达最高精度，下到最快速度，Scaled-YOLOv4：模型缩放显神威](https://mp.weixin.qq.com/s/Uo5pge7uq-Bh_wR9Vd5M7w)\n   * :star:[Github](https://github.com/WongKinYiu/ScaledYOLOv4) \n* YOLOv5\n  * :star:[code](https://github.com/ultralytics/yolov5)\n  * 解读：[YOLOv5来了！基于PyTorch，体积比YOLOv4小巧90%，速度却超2倍](https://mp.weixin.qq.com/s/QGPZQN4-nAMONtIrWQzBfQ)  \n* YOLObile\n  * :star:[code](https://github.com/nightsnack/YOLObile)\n  * [手机端 19FPS 的实时目标检测算法：YOLObile](https://mp.weixin.qq.com/s/8kyQoxvRPrU48xSJwvUGJA)\n  * [YOLObile:面向移动设备的「实时目标检测」算法](https://mp.weixin.qq.com/s/OP5iLZtIABNcn_LFyBWOeA)\n  \n  \n   \u003ca name=\"3\"/\u003e\n   \n   ## 1.NanoDet\n   \n* 解读：[NanoDet：轻量级（1.8MB）、超快速（移动端97fps）目标检测项目](https://mp.weixin.qq.com/s/KC-QxYZf2471OICDFra7Zw)\n   \n* 解读：[YOLO之外的另一选择，手机端97FPS的Anchor-Free目标检测模型NanoDet现已开源~](https://mp.weixin.qq.com/s/7mHhltqDcnYZdHWoRS_EBg)\n   * 1.8MB超轻量，跑的比 YOLO 快\n   * :star:[Github](https://github.com/RangiLyu/nanodet)\n\n\n# Salient Object Detection (SOD)\n\n\u003ca name=\"2\"/\u003e\n\n## 2.U^2-Net\n\n* 📜 [U2-Net: Going Deeper with Nested U-Structure for Salient Object Detection](https://arxiv.org/abs/2005.09007)\n   * ICPR 2020\n   * :star:[Github](https://github.com/NathanUA/U-2-Net)\n\n\u003ca name=\"1\"/\u003e\n\n## 1.BASNet\n\n* 📜 [BASNet: Boundary-Aware Salient Object Detection](https://openaccess.thecvf.com/content_CVPR_2019/html/Qin_BASNet_Boundary-Aware_Salient_Object_Detection_CVPR_2019_paper.html)\n   * 在单个GPU上的运行速度超过25 fps\n   * CVPR 2019\n   * :star:[Github](https://github.com/NathanUA/BASNet)\n\n\u003ca name=\"*\"/\u003e\n\n##  用于手机端实时检测 \n\n* NanoDet\n  * :star:[code](https://github.com/RangiLyu/nanodet)\n  * 解读：[NanoDet：轻量级（1.8MB）、超快速（移动端97fps）目标检测项目](https://mp.weixin.qq.com/s/KC-QxYZf2471OICDFra7Zw)\n  * 解读：[YOLO之外的另一选择，手机端97FPS的Anchor-Free目标检测模型NanoDet现已开源~](https://mp.weixin.qq.com/s/7mHhltqDcnYZdHWoRS_EBg)\n\n* YOLObile\n  * :star:[code](https://github.com/nightsnack/YOLObile)\n  * [手机端 19FPS 的实时目标检测算法：YOLObile](https://mp.weixin.qq.com/s/8kyQoxvRPrU48xSJwvUGJA)\n  * [YOLObile:面向移动设备的「实时目标检测」算法](https://mp.weixin.qq.com/s/OP5iLZtIABNcn_LFyBWOeA)\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F52cv%2Fobjectdetection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2F52cv%2Fobjectdetection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2F52cv%2Fobjectdetection/lists"}