Skip to content

Apply Deep learning models for Computer Vision using Streamlit

Notifications You must be signed in to change notification settings

amineHY/computervision-dashboard

Repository files navigation

🚀 Computer Vision Dashboard 🚀


Table of content

Docker image


GitHub URL

Link to this GitHub Repo.


Features

  • Apply Deep learning models for Computer Vision
  • Support for images and videos
  • Loading data from different source: local, web (URL), upload
  • It is designer with a modular architecture, where each app is a class
  • Export Analytics KPI to a CSV file
  • Display Analytics
  • Backend and Frontend separated and shipped in their respective docker image
  • Librairies: OpenCV, Python, pandas, Tensorflow, Streamlit
    • Frontend developed with python and streamlit
    • Backend developed in python with FastAPI, OpenCV, Tensorflow...

Architecture

Add and image architecture HERE

image


Demo

Video Applications

image


Image Applications


Launch the Dashboard

Run the dashboard from source

Prepare Python virtual environnement

  • Create a python virtual environnement using requirements.txt

    pipenv install -r requirements.txt

    Note the path for the created folder venv_folder image

  • Activate the environnement

    source venv_folder/bin/activate

    or

    pipenv shell

Run the dashboard

  • First run this command from the terminal

    streamlit run main.py

    image


Run the dashboard from docker

(Optional) Build the docker image

  • Build docker image
docker build -t aminehy/computervision-dashboard:latest .
  • Push the docker image to Docker Hub
docker login
docker push aminehy/computervision-dashboard:latest

Run the dashboard

docker run -it --rm aminehy/computervision-dashboard:latest streamlit run main.py --server.port 8050

image

About

Apply Deep learning models for Computer Vision using Streamlit

Topics

Resources

Stars

Watchers

Forks

Packages

No packages published