YOLO v2 Bounding Box Tool

Bounding box labeler tool to generate the training data in the format YOLO v2 requires.

Table of contents

• Quick start
• Prerequisites
• Run project
• GUI usage
• Authors

Quick start

To start using the YOLO Bounding Box Tool you need to download the latest release or clone the repo:

git clone https://github.com/Cartucho/yolo-boundingbox-labeler-GUI

Prerequisites

You need to install:

• OpenCV version >= 3.0
  • Installation in Windows
  • Installation in Linux
• Python

Run project

Step by step:

1. Insert the images in the folder images/
2. Insert the class list in the file class_list.txt
3. Run the code: python run.py
4. You can find the bounding box files in the folder bbox_txt/

GUI usage

Keyboard, press:

Key	Description
h	help
q	quit
e	edges
a/d	previous/next image
s/w	previous/next class

Mouse:

• Use two left clicks to do each bounding box
• Use the middle mouse to zoom in and out

Authors

• João Cartucho - Please give me your feedback: to.cartucho@gmail.com

  Feel free to contribute