Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Contribute to PaddleDetection] Yes, PP-YOLOE! 基于PP-YOLOE的算法开发 #7345

Open
thinkthinking opened this issue Nov 16, 2022 · 8 comments
Labels
contributor help wanted Extra attention is needed

Comments

@thinkthinking
Copy link
Collaborator

thinkthinking commented Nov 16, 2022

[Contribute to PaddleDetection] Yes, PP-YOLOE! 基于PP-YOLOE的算法开发

本期活动联系人:thinkthinking

建设目标

PP-YOLOE+是百度飞桨团队开源的最新SOTA通用检测模型,COCO数据集精度达54.7mAP,其L版本相比YOLOv7精度提升1.9%,V100端到端(包含前后处理)推理速度达42.2FPS。

我们鼓励大家基于PP-YOLOE去做新的算法开发,比如:

  • 改造PP-YOLOE适用于旋转框、小目标、关键点检测、实例分割等场景;
  • 精调PP-YOLOE用于工业质检、火灾检测、垃圾检测等垂类场景;
  • 将PP-YOLOE用于PP-Human、PP-Vehicle等Pipeline中,提升pipeline的检测效果。

相信通过这些活动,大家可以对PP-YOLOE的细节有更深刻的理解,对业务场景的应用也可以做更细节的适配。

参与方式

  • 方式一列表选题,见招募列表(提供了选题方向、题目、优秀的对标项目、文章和代码,以供学习)。
  • 方式二自选题目,对于非参考列表内的题目,可自主命题,需要与负责人 thinkthinking讨论后决定题目。

题目认领

为避免重复选题、知晓任务状态、方便统计管理,请根据如下操作认领您的题目。

在本issue提交题目:issue

  • 方式一(列表选题):在“招募列表”中选择题目,并在issue中,回复下列信息:

【列表选题】
编号:XX
题目:XXXX
认领人:XX
  • 方式二(自选题目):自主命题,直接在 issue 中,回复下列信息:
【自选题目】
题目:XXXX
认领人:XX

招募列表

序号 类型 题目 难度 参考 认领人
01 模型改造 PP-YOLOE用于旋转框检测 https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.5/configs/rotate PaddleDetection
02 模型改造 PP-YOLOE用于小目标检测 https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.5/configs/smalldet DrRyanHuang, PaddleDetection
03 模型改造 PP-YOLOE用于关键点检测 https://github.com/WongKinYiu/yolov7/tree/pose zetabao
04 模型改造 PP-YOLOE用于实例分割 https://github.com/WongKinYiu/yolov7/tree/mask ----
05 垂类应用 基于PP-YOLOE的缺陷检测 https://aistudio.baidu.com/aistudio/projectdetail/2367089 ----
06 垂类应用 基于PP-YOLOE的行为检测 https://aistudio.baidu.com/aistudio/projectdetail/2500639 ----
07 垂类应用 基于PP-YOLOE的异物检测 https://aistudio.baidu.com/aistudio/projectdetail/3846170?channelType=0&channel=0 ----
08 垂类应用 基于PP-YOLOE的安全监测 https://aistudio.baidu.com/aistudio/projectdetail/2503301?channelType=0&channel=0 ----
09 Pipeline PP-YOLOE-->PP-Human大升级 https://aistudio.baidu.com/aistudio/projectdetail/4606001 ----
10 Pipeline PP-YOLOE-->PP-Vehicle大升级 https://aistudio.baidu.com/aistudio/projectdetail/4512254 ----

【注意】招募列表外的,欢迎开发者联系活动负责人thinkthinking提交贡献👏

贡献指南

  1. 提ISSUE、PR的步骤请参考飞桨官网-贡献指南-代码贡献流程
  2. AI-Studio使用指南请参考AI-Studio新手指南

原则及注意事项

  1. 需使用PaddlePaddle框架, 建议复用PaddleDetection代码。
  2. 建议使用Paddle框架最新版本.
  3. PR需提到PaddleDetection-develop分支。
  4. 模型改造类的任务建议以PR形式提交
  5. 垂类应用以及Pipeline类的任务建议以AI-Studio项目形式提交,项目会同步到产业范例页面

还有不清楚的问题

欢迎大家随时在本issue下提问,飞桨会有专门的管理员进行疑问解答。

有任何问题,请联系本期活动联系人 thinkthinking

非常感谢大家为飞桨贡献!共建飞桨繁荣社区!

@thinkthinking thinkthinking pinned this issue Nov 16, 2022
@thinkthinking thinkthinking added help wanted Extra attention is needed contributor labels Nov 18, 2022
@DrRyanHuang
Copy link
Member

【列表选题】
编号:02
题目:PP-YOLOE用于小目标检测
认领人:Ryan

@thinkthinking
Copy link
Collaborator Author

【列表选题】 编号:02 题目:PP-YOLOE用于小目标检测 认领人:Ryan

赞!已更新招募列表~

@zetabao
Copy link

zetabao commented Dec 12, 2022

【列表选题】
编号:03
题目:PP-YOLOE用于关键点检测
认领人:YaoYao

@thinkthinking
Copy link
Collaborator Author

【列表选题】 编号:03 题目:PP-YOLOE用于关键点检测 认领人:YaoYao

@zetabao 收到!欢迎参与!👏

@daoqiugsy
Copy link

活动现在还在进行吗?

1 similar comment
@daoqiugsy
Copy link

活动现在还在进行吗?

@thinkthinking
Copy link
Collaborator Author

活动现在还在进行吗?
@daoqiugsy 你好,此活动长期都在进行,欢迎参与!

@thinkthinking thinkthinking unpinned this issue Apr 26, 2023
@qxfoxin
Copy link

qxfoxin commented Apr 22, 2024

你好,请问PP-Human检测跟踪模块的行人闯入分析的数据可以怎么导出,将检测到的总人数、闯入区域的人数及在区域中停留的时间导出到csv

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
contributor help wanted Extra attention is needed
Projects
None yet
Development

No branches or pull requests

5 participants