IMAGE CAPTIONING USING VIT-ENCODER AND GPT2 DECODER

1. Clone this repository.

2. Change directory using (cd ~/Image-Captioning-GO)

4. Run 'pip install -r requirements.txt'

5. Go to the 'caption-app' directory and run npm install and npm run start(if this throws error, try installing react-scripts directly:

'npm install react-scripts --save-dev' and 'npm run start'). A react app will run on port 3000 and open in default browser, allow camera access.

6. Go to the project root directory and run 'go run main.go'. A back-end service will run.

7. Staying in the root directory run source venv/bin/activate then again run 'python3 app.py'. A python app for image captioning will run.(run pip install -r requirements.txt to install python dependencies)

8. Your browser now produces caption for images captured using webcam.

##ARCHITECTURE