🧠 Visual Memory Search

A powerful tool to search your screenshot history using natural language, understanding both text and visual content.

---

✨ Features

• 🧠 Hybrid Search: Combines AI analysis of image pixels and OCR text for highly accurate results.
• 🖥️ Intuitive GUI: A simple and clean web interface built with Streamlit for easy searching.
• ⌨️ Powerful CLI: A full-featured command-line interface for scripting and power users.
• 🤖 Automated Indexing: A background "watcher" automatically detects and indexes new screenshots.
• 🔒 Local First: All models and data are processed locally on your machine. Nothing is sent to the cloud.

---

🚀 Getting Started

Follow these steps to get the project running on your local machine.

Prerequisites

• Python 3.9 or higher
• Git

Installation

1. Clone the Repository:

   git clone https://github.com/Ahmed-Rahil/Visual-Memory-Search.git
   
   cd visual-memory-search

2. Create and Activate a Virtual Environment:

   # macOS / Linux
   python3 -m venv .venv
   source .venv/bin/activate
   
   # Windows
   python -m venv .venv
   .\.venv\Scripts\activate

3. Install Dependencies:

   pip install -r requirements.txt

4. Add Your Screenshots: Place any images you want to index into the screenshots/ folder.

---

🖥️ How to Use

You can run this project in two ways. The GUI is recommended for the best experience.

GUI Application (Recommended)

This method requires two separate terminals running simultaneously.

1. Start the Automated Watcher (Terminal 1): This script monitors your screenshots folder and automatically indexes new images. Keep it running in the background.

   python watcher.py

2. Launch the GUI (Terminal 2): This will start the Streamlit web server and open the application in your browser.

   streamlit run app.py

Command-Line Interface (CLI)

For scripting or terminal-based workflows, use the unified CLI.

• Index a Folder (One-Time Scan):

  python vms/main.py index --path /path/to/screenshots

• Search from the Terminal:

  python vms/main.py search "a python function with a for loop"

• Watch a Folder (CLI version):

  python vms/main.py watch

---

🛠️ Configuration

Project settings can be modified in the config.yaml file:

• embedding_model: The AI model to use.
• database: The path for the vector database.
• default_screenshots_folder: The default folder to index and watch.

Contibution

Contribution is most welcome. Create an Issue if you have suggestion and create a pull request if you have some contibution.