import gradio as gr
import cv2
import numpy as np
from deep_sort_realtime.deepsort_tracker import DeepSort

# Load YOLO model and configuration
net = cv2.dnn.readNet("yolov3.weights", "yolov3.cfg")
with open("coco.names", "r") as f:
    classes = [line.strip() for line in f.readlines()]

# Initialize DeepSORT tracker
tracker = DeepSort(max_age=30, n_init=3, nn_budget=20)

def count_unique_people(video_path):
    # Open video
    cap = cv2.VideoCapture(video_path)
    if not cap.isOpened():
        return "Error: Unable to open video file."
    
    unique_people = set()  # To store unique IDs
    frame_count = 0

    while cap.isOpened():
        ret, frame = cap.read()
        if not ret:
            break

        frame_count += 1
        height, width, _ = frame.shape

        # Detect people using YOLO
        blob = cv2.dnn.blobFromImage(frame, 1 / 255.0, (416, 416), swapRB=True, crop=False)
        net.setInput(blob)
        output_layers_names = net.getUnconnectedOutLayersNames()
        layer_outputs = net.forward(output_layers_names)

        boxes = []
        confidences = []
        for output in layer_outputs:
            for detection in output:
                scores = detection[5:]
                class_id = np.argmax(scores)
                confidence = scores[class_id]

                # If detected class is 'person'
                if classes[class_id] == "person" and confidence > 0.5:
                    center_x = int(detection[0] * width)
                    center_y = int(detection[1] * height)
                    w = int(detection[2] * width)
                    h = int(detection[3] * height)
                    x = int(center_x - w / 2)
                    y = int(center_y - h / 2)
                    boxes.append([x, y, w, h])
                    confidences.append(float(confidence))

        # Apply non-maximum suppression
        indexes = cv2.dnn.NMSBoxes(boxes, confidences, 0.5, 0.4)
        detections = []
        if len(indexes) > 0:
            for i in indexes.flatten():
                x, y, w, h = boxes[i]
                detections.append(([x, y, x + w, y + h], confidences[i]))

        # Update tracker with detections
        tracks = tracker.update_tracks(detections, frame=frame)

        # Track unique IDs
        for track in tracks:
            if not track.is_confirmed():
                continue
            track_id = track.track_id
            unique_people.add(track_id)

    cap.release()

    return {
        "Total Unique People Detected": len(unique_people),
        "Total Frames Processed": frame_count,
    }

# Gradio Interface
description = """
Upload a video, and the app will count the total number of unique people detected in the video using YOLO and DeepSORT.
"""
interface = gr.Interface(
    fn=count_unique_people,
    inputs=gr.Video(label="Upload Video"),
    outputs=gr.JSON(label="Unique People Count"),
    title="Unique People Counter",
    description=description,
)

if __name__ == "__main__":
    interface.launch(server_name="0.0.0.0", server_port=7860)