import gradio as gr
import cv2
import requests
import os
from ultralytics import YOLO

# URLs of files to be downloaded
file_urls = [
    'https://www.dropbox.com/s/b5g97xo901zb3ds/pothole_example.jpg?dl=1',
    'https://www.dropbox.com/s/86uxlxxlm1iaexa/pothole_screenshot.png?dl=1',
    'https://www.dropbox.com/s/7sjfwncffg8xej2/video_7.mp4?dl=1'
]

# Function to download a file if it does not exist locally
def download_file(url, save_name):
    if not os.path.exists(save_name):
        response = requests.get(url)
        with open(save_name, 'wb') as file:
            file.write(response.content)

# Download files using the URLs
for i, url in enumerate(file_urls):
    save_name = "video.mp4" if url.endswith('.mp4') else f"image_{i}.jpg"
    download_file(url, save_name)

# Load the YOLO model
model = YOLO('best.pt')

# Paths for example images and video
image_paths = [['image_0.jpg'], ['image_1.jpg']]
video_path = [['video.mp4']]

# Function to show predictions on an image
def show_preds_image(image_path):
    image = cv2.imread(image_path)
    outputs = model.predict(source=image_path)
    results = outputs[0].cpu().numpy()
    for det in results.boxes.xyxy:
        cv2.rectangle(
            image,
            (int(det[0]), int(det[1])),
            (int(det[2]), int(det[3])),
            color=(0, 0, 255),
            thickness=2,
            lineType=cv2.LINE_AA
        )
    return cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

# Gradio interface for image inference
inputs_image = gr.Image(type="filepath", label="Input Image")
outputs_image = gr.Image(type="numpy", label="Output Image")
interface_image = gr.Interface(
    fn=show_preds_image,
    inputs=inputs_image,
    outputs=outputs_image,
    title="Pothole Detector - Image Inference",
    examples=image_paths,
    cache_examples=False,
)

# Function to show predictions on a video
def show_preds_video(video_path):
    cap = cv2.VideoCapture(video_path)
    while cap.isOpened():
        ret, frame = cap.read()
        if not ret:
            break
        frame_copy = frame.copy()
        outputs = model.predict(source=frame)
        results = outputs[0].cpu().numpy()
        for det in results.boxes.xyxy:
            cv2.rectangle(
                frame_copy,
                (int(det[0]), int(det[1])),
                (int(det[2]), int(det[3])),
                color=(0, 0, 255),
                thickness=2,
                lineType=cv2.LINE_AA
            )
        yield cv2.cvtColor(frame_copy, cv2.COLOR_BGR2RGB)
    cap.release()

# Gradio interface for video inference
inputs_video = gr.Video(type="filepath", label="Input Video")
outputs_video = gr.Image(type="numpy", label="Output Frame")
interface_video = gr.Interface(
    fn=show_preds_video,
    inputs=inputs_video,
    outputs=outputs_video,
    title="Pothole Detector - Video Inference",
    examples=video_path,
    cache_examples=False,
)

# Combine the image and video interfaces into a tabbed interface
gr.TabbedInterface(
    [interface_image, interface_video],
    tab_names=['Image Inference', 'Video Inference']
).queue().launch()