tinder_streamlit

In this world were it becomes hard to distinguish between real and generated images we have created a Tinder App where people can vote on whether the image is real or generated by an AI.

Welcome to our Tinder Streamlit App! This app is a web application built with Streamlit, a Python library for building web applications. It provides a simple and interactive way to explore and interact with images.

📦 Repo Structure

├── Home.py
├── README.md
├── animations
│   ├── Olga.mp4
│   ├── background.avif
│   ├── background2.webp
│   ├── bg3.jpg
│   ├── hourglass.json
│   ├── style.css
│   └── themes.json
├── arcade.db
├── images
│   ├── genai
│   │   ├── alice_1.png
│   │   ├── ...
│   │   ├── bear_1.png
│   │   ├── ...
│   │   ├── dar_1.png
│   │   ├── ...
│   │   ├── nathalie_1.png
│   │   └── ...
│   └── real
│       ├── pexel_1.jpg
│       └── ...
├── pages
│   ├── 1_🎮_Facemash_GAME.py
│   ├── 1_🎮_Tinder_GAME.py
│   ├── 2_⬆️_Contribute.py
│   └── 3_🏆_Leaderboard.py
├── requirements.txt
└── utils
    ├── Contribute_file.py
    ├── Database_file.py
    ├── Game_file.py
    ├── Home_file.py
    ├── Leaderboard_file.py
    ├── Theme_file.py
    ├── __init__.py
    └── __pycache__

🚀 Project flow chart:

flowchart TD
    X(Home page)
    A(CONTRIBUTE)
    B(Fill in password)
    L{pwd correct}
    C(select image or images)
    D(select AI or real)
    E(Uploads image to images folder)
    F(update database)

    G(GAME)
    H(fill in name)
    I(click on what image is most likely AI)
    J{Correct guess?}
    K(One image goes away one stays)
    N(update scores table)
    O(update scores table)
    M(LEADERBOARD)

    X --> A
    X --> G
    X --> M

    A --> B
    B --> L
    L -- YES --> D
    L -- NO --> B
    D --> C
    C --> E
    E --> F

    G --> H
    H --> I
    I --> J
    J --YES --> N
    J --NO --> O
    N --> K
    K --> I
    O --> M

Loading

🎮 Setup Instructions

Step 1: Clone repo

git clone git@github.com:bear-revels/tinder_streamlit.git

Step 2: Install dependencies

pip install -r requirements.txt

Step 3: Run streamlit app locally

streamlit run app.py

📦 Class diagrams

These diagrams were created using https://github.com/scottrogowski/code2flow and https://dreampuf.github.io/GraphvizOnline/

Creation of a graph can be done with the following command:

code2flow Home.py utils/*.py pages/*.py

The separate class diagrams are as follows:

The full flow of the project looks like this:

📌 Background

This team project was completed in 5 days as part of the AI Boocamp at BeCode.org. Connect with the team behind the magic.

1. Bear Revels
2. Niels Demeyer
3. Caroline Van Hoeke
4. Nathalie Nachtergaele

🙋‍♂️ Special kudos to the initial BANDY team

• for investigating opensource models
• for creating the best AI images possible
• for laying the foundations of this app/game

1. Alice Mendes
2. Bear Revels
3. Daryoush Ghanbarpour
4. Nathalie Nachtergaele
5. Yanina Andriienko