README.md

This is the offical repository of our project moni for the hackathon Hackathon Thurgau 2023. The project was created and developed during the hackathon competition.

The project aims to provide a solution for the Challenge 3: occupancy measurement of a store. This was achieved by using a combination of computer vision and a cnn model (for ReID) to detect and track people in a store and their movement. The data is then used for further analysis and visualizations.

Quicklinks

• Installation & Setup Guide
• Project Structure
  • Folder Structure
  • Mermaid Diagram
• Demo

Project Structure

Folder Structure

moni
├─── conf                               Configuration files
|    └─── example-config.yml            Example configuration file
├─── examples                           Configuration files
|    └─── homography.ipynb              Jupyter Notebook for homography show case
├─── github-content                     Images for the README.md
|    |─── tech.md                       Technical documentation how to use moni
|    └─── mermaid_chart.txt             Mermaid chart of the moni architecture
├─── Yolov7_StrongSORT_OSNet            Submodule from: mikel-brostrom
├─── .gitignore                         Gitignore file
├─── .gitmodules                        Gitmodules file
├─── Dockerfile                         Dockerfile to containerize moni
├─── README.md                          README.md
├─── main.py                            Moni main python script
├─── runner_utils.py                    Util functions which are used by moni
├─── runner.py                          Runner function which does the processing work
├─── requirements.txt                   Requirements file
└─── docker-compose.yml                 Docker compose file to spin up the whole moni platform

Flow/Process Diagram

Demo

This is a demo of the running application with 3 different views:

Video Source: EPFL Labs

Authors

• JohnTrunix
• melrog
• thomasglauser