app.py

import cv2
from ultralytics import YOLO
import numpy as np
import os
import gradio as gr

def fonk(video_path):
  
  model=YOLO("best.pt")
  cap=cv2.VideoCapture(video_path)

  frame_width = int(cap.get(3)) 
  frame_height = int(cap.get(4))
  size = (frame_width, frame_height)
  output_video = cv2.VideoWriter('filename.mp4',  
                         cv2.VideoWriter_fourcc(*'XVID'), 
                         10, size) 

  
  while True:
    ret, frame= cap.read()

    if ret!=True:
      break

    results= model(frame)
    for result in results:
      box=result.boxes

      x1, y1, x2, y2 = map(int, box.xyxy[0])
      print(x1, y1, x2, y2)
      frame= cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 255, 0), 2)
      output_video.write(frame)
      
      return output_video

demo = gr.Interface(fonk,
                    inputs= gr.Video(),
                    outputs=gr.Video(),
                    examples=["cow-video-cows-mooing-and-grazing-in-a-field.mp4"],
                    title= "cows",
                    cache_examples=True)
demo.launch()