Giter Club home page Giter Club logo

centernet's Introduction

CenterNet

使用CenterNet来训练其他数据。

本项目实现以下

  • Face 人脸bbox+landmark检测
  • SeaShips 船只种类和bbox检测

代码参考以下

Objects as Points

CenterMulti

CenterPose

结果

Face

测试:使用原图分辨率,未使用Multi Scale,在验证集上进行测试。

Backbone AP AP50 AP75 APS APM APL
DLA-34 43.7 73.7 47.7 31.9 58.8 64.0
MobileNet-V2 34.3 62.0 35.0 27.2 52.4 42.7

img

img

SeaShips

测试:使用原图分辨率,未使用Multi Scale,在验证集和测试集上进行测试。

Backbone 数据集 AP AP50 AP75 APS APM APL
DLA-34 Val 21.3 37.3 21.5 10.8 39.3 22.1
DLA-34 Test 21.6 36.4 23.5 2.7 32.4 22.8

img

训练和测试

centernet's People

Contributors

hibugs avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar

centernet's Issues

confidence score

Thanks for your work!!!
I use this repo to detec my own plate dataset,when I finally get my train model ,this result is just ok ,but the confidence score is usually between from 0.4-0.8,for dectec task this score isn't too low?when I use retinanet for detection task ,the score is usually over 0.95,so can you gave me some advices on how to make the confidence score higher?

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.