app.py

import cv2 as cv
import numpy as np
import gradio as gr

def nostalgic_effect(image):
    # Convert the image to a NumPy array
    image = np.array(image)
    # Convert from RGB to BGR (Gradio inputs in RGB, OpenCV works in BGR)
    image = cv.cvtColor(image, cv.COLOR_RGB2BGR)
    # Convert to grayscale
    gray_image = cv.cvtColor(image, cv.COLOR_BGR2GRAY)
    return gray_image

# Creating the Gradio interface
with gr.Blocks() as demo:
    gr.Markdown("# Convert Image to Black and White!")
    gr.Markdown("Upload an image and see it transformed to black and white...")
    
    image_input = gr.Image(type='pil')
    image_output = gr.Image(type="numpy", label="Converted Image")
    
    # Connecting the function to Gradio components:
    convert_btn = gr.Button("Convert")
    convert_btn.click(fn=nostalgic_effect, inputs=image_input, outputs=image_output)

if __name__ == "__main__":
    demo.launch()