Text-Search-Image

This project implements an image search system using CLIP (Contrastive Language-Image Pre-training) model. It allows users to search for images using natural language queries.

Features

• Load and process images from a specified folder
• Encode images using CLIP model
• Store image embeddings in a vector database (ChromaDB)
• Search for images using text queries
• Display search results in a user-friendly interface

Requirements

• Python 3.9.16
• Flet: For creating the graphical user interface
• Clip Model: For generating document embeddings
• ChromaDB: For efficient storage and retrieval of vector embeddings

Installation

1. Clone this repository:

git clone https://github.com/PongpreechaSuea/Text-Search-Image.git
cd Text-Search-Image

2. Create a virtual environment (optional but recommended):

python -m venv test
source test/bin/activate  # On Windows, use `test\Scripts\activate`

3. Install the required packages:

pip install -r requirements.txt

Usage

1. Run the application:

python run.py

2. In the application interface:

   • Enter the path to the folder containing your images
   • Click the "Add" button to load and process the images
   • Enter a text query in the search box
   • Click the "Search" button to find matching images

Project Structure

text_search_image_clip/
├── assets/
├── data/
├── src/
│   ├── config.py
│   ├── image_finder.py
│   └── image_generator.py
├── README.md
├── requirements.txt
└── run.py

Example

DATA EAMPLE

INTERFACE

RESULTS