Skip to content

Use CNN techs (mainly YoLo and Farster RCNN) to generate front-end code by given web page screenshots or design scratch -[DOING]-

Notifications You must be signed in to change notification settings

zhangyonglei/Research-ReverselyGeneratingWebCode

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 

Repository files navigation

Researech-ReverselyGeneratingWebCode-Selenium

Attempt to generate layout code (HTML/CSS) from given web page screenshot [-ING]

First Stage


Preprocessing:

Gather data and building the training data set
- Crawl and collect webpage + Info
- Label component on the web page w.r.t the Info gathered
- Clip the components for classification

Training (Neural Network):

Build model to recognize the tag class
- Decide and build an appropriate model
- Fine tune

Transform (Image Processing):

Transform the input component images into HTML code
- Calculate the shape and size of components
- Calculate the color theme of components
- Classify the tag of components

About

Use CNN techs (mainly YoLo and Farster RCNN) to generate front-end code by given web page screenshots or design scratch -[DOING]-

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 77.2%
  • HTML 19.5%
  • CSS 2.1%
  • Cuda 0.8%
  • Lua 0.4%
  • C++ 0.0%