🔍 AI Image Search App with CLIP + Flask

An AI-powered Image Search App built with OpenAI's CLIP model and Flask.
Enter a text query like "a cute cat" and retrieve the most relevant images from your dataset.

---

🚀 Demo

---

✨ Features

• CLIP Embeddings: Encodes both text and images into the same semantic space.
• Text-to-Image Search: Finds the most relevant images using cosine similarity.
• Flask Web Interface: A simple search form with real-time results.
• CPU-Friendly: Works on any machine without GPU acceleration.

---

📦 Installation

1. Clone the Repository

git clone https://github.com/datageekrj/Flask-Image-Search-YouTube-Tutorial.git
cd Flask-Image-Search-YouTube-Tutorial

2. Install Dependencies

pip install torch torchvision pillow ftfy regex tqdm flask
pip install git+https://github.com/openai/CLIP.git

---

🖼️ Add Your Images

Place your dataset images inside:

static/images/

Example:

static/images/
  ├── cat1.jpg
  ├── dog1.jpg
  └── car1.jpg

---

▶️ Run the App

python app.py

Now open your browser at: http://127.0.0.1:5000

---

📂 Project Structure

clip-image-search/
│
├── app.py                   # Main Flask app
├── search.py                # Search logic using CLIP
├── generate_embeddings.py   # Precompute embeddings
├── templates/
│   └── index.html           # Frontend template
└── static/
    └── images/              # Image dataset

---

🎥 YouTube Tutorial

I’ve created a step-by-step video guide on building this project: 👉 https://youtu.be/38LsOFesigg

---

🧠 Next Steps / Improvements

• Add image-to-image search (query with an image instead of text).
• Use FAISS for fast vector search on large datasets.
• Deploy the app on Render, Vercel, or Heroku.

---

📜 License

This project is licensed under the MIT License – feel free to use and modify.

---

⭐ Show Your Support

If this project helped you, star the repo ⭐ and share it with others!