Test Image Retrieval

This project demonstrates content-based image retrieval using Streamlit and ChromaDB.

Project Setup

Clone the Repository

Clone the project from GitHub:

git clone https://github.com/NguyenHuy190303/Images-Retrieval-Streamlit

Navigate to the Project Directory

cd Images-Retrieval-Streamlit

Install Dependencies

Install the required dependencies using pip:

pip install -r requirements.txt

Setup and Running create_images_db.py

This script is used to set up the ChromaDB and process the training dataset.

1. Set Up Directory Paths

Before running the script, ensure you have the following directories set up on your system:

• Training Dataset: Store your images in the directory specified by the ROOTS variable.
• Database Path: The ChromaDB database will be stored in the directory specified by the db_path variable.

2. Update the Script

Open create_images_db.py and update the following variables:

# Path to your training dataset
ROOTS = r"D:\Python\Text_Image_Retrieval-Streamlit\data\train"

# Path to where the ChromaDB will be created/stored
db_path = r"D:\Python\Text_Image_Retrieval-Streamlit\database"

3. Initialize and Embed Vectors in ChromaDB

Run the script to initialize and embed vectors in the ChromaDB:

python create_images_db.py

Run Streamlit Application

Once the database is set up, you can run the Streamlit application:

streamlit run main.py