import gradio as gr
import numpy as np
import time

# Renamed for clarity and consistency
def fake_diffusion_denoise(image, steps):
    if image is None:
        yield np.zeros((100, 100, 3), dtype=np.uint8)
        return
    original_image = image.astype(np.float32) / 255.0
    # Add initial noise
    noisy_start_image = original_image + np.random.normal(0, 0.7, original_image.shape)
    noisy_start_image = np.clip(noisy_start_image, 0, 1)

    for i in range(steps):
        time.sleep(0.2)
        # Simulate denoising: gradually revert to the original image (linear progress)
        progress = (i + 1) / steps
        denoised_step = (1 - progress) * noisy_start_image + progress * original_image
        denoised_step = np.clip(denoised_step, 0, 1)
        yield (denoised_step * 255).astype(np.uint8)
    yield (original_image * 255).astype(np.uint8) # Ensure final image is clean

def real_diffusion_add_noise(image, steps):
    if image is None:
        yield np.zeros((100, 100, 3), dtype=np.uint8)
        return
    base_image = image.astype(np.float32) / 255.0
    max_noise_std = 0.8 # Maximum noise level to reach

    for i in range(steps):
        time.sleep(0.2)
        # Increase noise progressively
        current_noise_std = max_noise_std * ((i + 1) / steps)
        noise = np.random.normal(0, current_noise_std, base_image.shape)
        noisy_step = base_image + noise
        noisy_step = np.clip(noisy_step, 0, 1)
        yield (noisy_step * 255).astype(np.uint8)
    # Yield the most noisy version as the final step
    final_noise = np.random.normal(0, max_noise_std, base_image.shape)
    final_noisy_image = np.clip(base_image + final_noise, 0, 1)
    yield (final_noisy_image * 255).astype(np.uint8)


def flow_matching_denoise(image, steps):
    if image is None:
        yield np.zeros((100, 100, 3), dtype=np.uint8)
        return
    original_image = image.astype(np.float32) / 255.0
    # Start with a significantly noisy image
    very_noisy_image = original_image + np.random.normal(0, 1.0, original_image.shape) # High initial noise
    very_noisy_image = np.clip(very_noisy_image, 0, 1)

    for i in range(steps):
        time.sleep(0.2)
        # Non-linear progress using a sigmoid-like curve for smoother transition
        p_norm = (i + 1) / steps  # Normalized progress 0 to 1
        # Transform p_norm to a range like -5 to 5 for sigmoid
        sigmoid_input = 10 * (p_norm - 0.5)
        flow_progress = 1 / (1 + np.exp(-sigmoid_input))

        denoised_step = (1 - flow_progress) * very_noisy_image + flow_progress * original_image
        denoised_step = np.clip(denoised_step, 0, 1)
        yield (denoised_step * 255).astype(np.uint8)
    yield (original_image * 255).astype(np.uint8) # Ensure final image is clean

# Main processing function that routes to different methods
def process_image_selected_method(method_selection, input_image, num_steps):
    if input_image is None:
        # This case should ideally be handled by Gradio if a default image URL is provided
        # or prevented by making the image input mandatory.
        # Yielding a placeholder if it somehow becomes None during processing.
        yield np.zeros((200, 200, 3), dtype=np.uint8)
        return

    if method_selection == "Fake Diffusion (Denoise)":
        yield from fake_diffusion_denoise(input_image, num_steps)
    elif method_selection == "Real Diffusion (Add Noise)":
        yield from real_diffusion_add_noise(input_image, num_steps)
    elif method_selection == "Flow Matching (Denoise)":
        yield from flow_matching_denoise(input_image, num_steps)
    else:
        # Default behavior: return the original image as is, or an error image
        yield input_image


method_choices = ["Fake Diffusion (Denoise)", "Real Diffusion (Add Noise)", "Flow Matching (Denoise)"]

demo = gr.Interface(
    fn=process_image_selected_method, # Use the router function
    inputs=[
        gr.Dropdown(choices=method_choices, label="Select Method", value="Fake Diffusion (Denoise)"),
        gr.Image(type="numpy", label="Input Image", value="https://gradio-builds.s3.amazonaws.com/diffusion_image/cute_dog.jpg"),
        gr.Slider(minimum=1, maximum=30, value=10, step=1, label="Processing Steps")
    ],
    outputs=gr.Image(type="numpy", label="Processed Image"),
    title="Diffusion Processing Demo",
    description="Select a method: 'Fake Diffusion (Denoise)' and 'Flow Matching (Denoise)' will denoise an image. 'Real Diffusion (Add Noise)' will progressively add noise to the image. Adjust steps for granularity."
)

# define queue - required for generators
demo.queue()

demo.launch()