Project: Implementation of Label Editor

 This is second article in Python Projects Series. This article includes the explanation and instructions of demo of this project. We also touch up on the topics like design and implementation of this project.

This Python code works as a label editor through which you can draw bounding boxes and label them. CSV file shall be generated from this tool which contains information about your labelling.

In case if you are new to this series, Please visit our previous article in this series.

1. Project : Implementation of Block Jump Game

2. Project: Implementation of Car Lane Change Game

 

In case if you are new to Python, Please visit the following articles in series Python for Data Science.

  1. Python for Data Science Series - Part 1
  2. Python for Data Science Series - Part 2
  3. Using Numpy in Python
  4. Using Pandas in Python
  5. Data Visualization using Matplotlib Python
  6. Parsing XML file in python
  7. Parsing Json file in python
  8. Interactive Data Visualization using Plotly in Python
  9. Logging in Python
  10. Read and Write Images in Python

1. Setup Project Environment:

    Code can be obtained from our git repository by using following command

git clone https://github.com/RajuSaladi/LabelEditor.git

    Go to "LabelEditor" Folder using following command

cd LabelEditor

    This Project requires the following python libraries to be installed. Please install them using command below.

  • OpenCV  pip install opencv-contrib-python
  • Numpy    pip install numpy
  • Pandas  pip install pandas
Setup folders in this folder sturcture: 
MainFolder
Data 
GroundTruth 
GroundTruthInfoMutliLabel.csv 
Labels.txt

Data: This folder should contain images to be labelled 
GroundTruth: This Folder to save the outputs of label tool 
Labels.txt: This text file should contain the label names for the data 
GroundTruthInfoMutliLabel.csv : Generated .csv file contains the Image Path , Object ID,ObjectType,x,y,w,h

2. Run Project:

Project can be run using following command

python MultiLabelEditor.py

When run successfully, it will popup a window which opens images that are placed in the folder "data" one after the other. User can draw bounding box on image using mouse left button i.e., Click and hold left mouse button where user want to start bounding box.move mouse towards the end of object. user can see the bounding box being drawn. when the user satisfies with bounding box drawn, release mouse left button.
A Popup will appear from which user can chose label for that bounding box
User can draw multiple bounding boxes
Once, labeling for the frame is done, Press 'd' on Keyboard.
If user wants to reset the bounding boxes for this frame, press 'r'
If user wants to skip the frame, press 'c'

Information of the Bounding Box shall be stored in the GroundTruthInfoMutliLabel.csv

Detailed explanation about this Project can be seen in the following video




(TO BE CONTINUED)

Stay Tuned for more information

Comments

Post a Comment

Popular posts from this blog

How to run Jupyter Notebooks in Cloud

 Hello all, In this article, I explained on how to open and run Jupyter notebooks in Cloud/Server. As most of cloud machines prefer Ubuntu or other Linux operating systems, we may only be able to access virtual machine in cloud using command line interface (CLI). In that case, it would be a challenge to open Jupyter notebooks as we usually access it through a browser. I would like to take an example of Azure VM for explaination but this would be valid for most of cloud platforms. Login to VM Use the following command to login to your VM thru ssh ssh username@ipaddress In case if you dont have a VM, you can create one by logging onto Azure Portal and following the steps mentioned in this Azure VM Documentation Using TMUX Tmux is very useful tool which will allow to run multiple terminal sessions within terminal and also let the code running even though you closed the terminal Use tmux ls for list of open tmux sessions Use tmux to create a new session Use tmux attach to go to alrea...
Read more

GUI Programming using Tkinter Python

Image
Hello all, This is the eleventh article in the series  Python for Data Science . If you are new to this series, we would recommend you to read our previous articles Python for Data Science Series - Part 1 Python for Data Science Series - Part 2 Using Numpy in Python Using Pandas in Python Data Visualization using Matplotlib Python Parsing XML file in python Parsing Json file in python Interactive Data Visualization using Plotly in Python Logging in Python Read and Write operations on Images using OpenCV Python Tkinter is Python library which can be used to make GUI Application in Python. This is available in python by default. In this article, we would explain the various functionalities available in Tkinter and also create a sample GUI Application which would look like below Please refer the videos below for detailed explanation on how to use Tkinter to create GUI elements in Python We have taken article from following webpage as reference Please refer the following code to unders...
Read more

How to download , install and configure Anaconda - Python

Image
Hello all, This article is aimed at explaining step by step procedure in installing and configuring Anaconda in Windows PC. Downloading Anaconda Latest Version 1. Go to the Anaconda web page by clicking the following link   https://www.anaconda.com/products/individual 2. If you scroll a little, you will find download button, please click on it. which will make the page to jump to the download section which would look like below. 3.  You will find various options to download. Please select an option depending on your operating system. 3.1. For Windows 64 bit, I would recommend downloading 64 Bit Graphical Installer. You click on that, this would start downloading  Anaconda3-2020.07-Windows-x86_64.exe file to your PC. This name will vary depending on the version which is latest by the time you open it. Installing Anaconda Latest Version: Once Download is finished, Please open the installer file and follow as per screenshots Click on "Next" Click on "I Agree" button i...
Read more