app.py

#!/usr/bin/env python
# coding: utf-8

# In[ ]:


# import required libraries
from ultralytics import YOLO
import gradio as gr
import cv2
import math
from src.items import classNames


# In[ ]:


# detection function
def yolo_detect(feed, vid):
    video = vid
    # Load a pretrained YOLOv8n model
    model = YOLO('yolov8n.pt')
    
    # Run inference on the source
    results = model(video, stream=True, verbose=False) 
    frames = list()
    
    # plot annotations
    for frame in results:
        boxes = frame.boxes
        single = frame.orig_img
        for box in boxes:
            # bounding box
            x1, y1, x2, y2 = box.xyxy[0]
            x1, y1, x2, y2 = int(x1), int(y1), int(x2), int(y2) # convert to int values

            # put box in cam
            cv2.rectangle(single, (x1, y1), (x2, y2), (255, 0, 255), 3)

            # object details
            cv2.putText(single, classNames[int(box.cls[0])], (x1,y1), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 0, 0), 1)
            
        frames.append(single)
    cv2.destroyAllWindows()
    
    h, w, c = frames[1].shape
    
    out_file = "output.avi"
    fourcc=cv2.VideoWriter_fourcc('X', 'V', 'I', 'D')
    writer = out = cv2.VideoWriter(out_file, fourcc, 25.0, (w, h))
    for i in range(len(frames)):
        writer.write(frames[i])
    writer.release()
    return out_file


# In[ ]:


demo = gr.Interface(fn=yolo_detect, 
                    inputs=[gr.PlayableVideo(source='webcam'), gr.Video(autoplay=True)],
                    outputs=[gr.PlayableVideo(autoplay=True, format='avi')],
                    cache_examples=True, allow_flagging='never')
demo.queue()
demo.launch(inline=False, debug=True, show_api=False, quiet=True)