lcydllcy / autolabelimg

A self automatically labeling tool

AutoLabelImg for Windows

**This is an experimental project , embedding mobilenet detector on labelimg

Windows Setup

Requirements

• Visual Studio 2013 or 2015
• CMake 3.4 or higher (Visual Studio and Ninja generators are supported)
• Anaconda (python2.7)
• PyQt4

Configuring and Building

> git clone https://github.com/eric612/AutoLabelImg.git --recursive
> conda install -c anaconda pyqt=4
> pip install -i https://pypi.anaconda.org/ales-erjavec/simple pyqt4
> (Option) pip install lxml 
> (Option) pip install opencv (or use anaconda navigator to install)
> (Option) pip install scikits-image (or use anaconda navigator to install)
> (Option) pip install protobuf

Download pre-train weights from original web and save at

$caffe_root\models\MobileNet\

> cd $caffe_root\script\
> build_win.cmd

Drink a coffe

> cd $AutoLabelImg_root\
> pyrcc4 -o resources.py resources.qrc

Usage

1. Python labelimg.py
2. click button "Change Save Dir" , and select your xml folder.
3. click button "Open Dir" and select image folder (example image : caffe/data/VOC0712)

demo video