Pixelio-Project

Overview

This repository houses a project that finds the top x images similar to any given photo from a large dataset. The software utilizes Python to recommend images based on various similarity metrics such as color schemes, embeddings, and YOLO object detection.

Getting Started

Prerequisites

Ensure Python is installed on your system and you have the necessary permissions to execute scripts.

Installation

Clone the repository to your local machine:

git clone https://github.com/AlhaririAnas/Pixelio-Project

Navigate to the project directory:

cd Pixelio-Project

Install dependencies:

pip install .

Setting up your dataset

To set up a dataset with all similarities and metadata, use:

python -m resources.main -m -s -p "../Parent_folder_of_dataset"

Use this commands to specify:

Command	Explanation	Tip
-m	set up the image_metadata.db
-s	set up a pickle file with all similarity information
-p	path to your '/data' directory with your images	Default: 'D:/'
-d	device to use: cpu or cuda	If not specified, cuda will be used if available
--pkl_file	path to a pickle file, in which the similarity information are stored	Default: 'similarities.pkl'
--checkpoint	Number of images after which the --pkl_file will be updated	Default: 100

If the program crashes, you can simply restart it and it will continue where it left off.

If neither the -m nor the -s flag are specified, it will launch the web interface...

Launching the Web Interface

To start the web-based interface, use the following command:

python -m resources.main --color

Make sure that the metadata.db file is located in the Pixelio project folder. The similarities.pkl should also be located there, or you can refer to it using --pkl_file.
The complete data set must be located in /app/static/images, e.g:

/app/
    /static/
        /images/
            /000/
                01.jpg
                02.jpg
            /001/
                03.png
                04.png
    /templates/
        index.html
    app.py

In addition, --image_count can be used to set how many similar images are to be calculated. Besides --color (color histograms) there are --embedding (Inception v3 embeddings) and --yolo (YOLO v8 object detection) as selection of the measuring method. An example could be:

python -m resources.main --yolo --image_count 30

After running the command, the system will open a browser window automatically. This process may take some time, especially if you are performing it for the first time. This is because the pickle file loads.

Pixelio Demo

If you click on an image, a sidebar will appear on the right side containing the most similar images to this image.

Contact

For any inquiries, please contact:

• Jonah Gräfe: jonah.graefe@study.hs-duesseldorf.de
• Joschua Schramm: joschua.schramm@study.hs-duesseldorf.de
• Anas Alhariri: anas.alhariri@study.hs-duesseldorf.de