AI-Image-Captioning

An AI-powered image captioning app built with Streamlit, using ViT-GPT2 for caption generation and YOLOv8 for object detection. The app enhances captions by integrating detected objects into the generated text.

🔥 Features

• AI-powered image captioning using ViT-GPT2.
• Object detection with YOLOv8 to enhance captions.
• Dark-themed UI with Streamlit.
• Interactive settings for enabling/disabling object detection.
• Optimized inference with GPU acceleration (CUDA support).

🚀 Demo

1️⃣ Upload an Image

2️⃣ Enable Object Detection and Generate Captions

3️⃣ View Enhanced Caption and Detected Objects

📂 Installation & Setup

1️⃣ Clone the Repository

git clone https://github.com/yourusername/AI-Image-Captioning.git
cd AI-Image-Captioning

2️⃣ Create a Virtual Environment (Optional but Recommended)

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

3️⃣ Install Dependencies

pip install -r requirements.txt

4️⃣ Run the Application

streamlit run app.py

🧠 Models Used

1️⃣ ViT-GPT2 (Image Captioning)

• Pretrained Model: nlpconnect/vit-gpt2-image-captioning
• Task: Generates textual descriptions for input images.

2️⃣ YOLOv8 (Object Detection)

• Pretrained Model: yolov8n.pt
• Task: Detects objects in the image to enhance captions.

⚙️ Project Structure

AI-Image-Captioning/
│── app.py                  # Main Streamlit application
│── requirements.txt        # Required dependencies
│── README.md               # Documentation
│── assest/                 # Store images/screenshots

🛠️ Usage Instructions

1. Upload an image in the app.
2. Choose whether to enable object detection.
3. Click 'Analyze Image' to generate a caption.
4. View enhanced captions and object detection results.

💡 Future Improvements

• Add multilingual captioning support.
• Optimize object detection performance.
• Implement additional caption refinement techniques.

🤝 Contributing

Contributions are welcome! Feel free to fork this repository and create a pull request with your improvements.