Create a Football Players Detection using YOLO v8 model and Open CV in Colab

I want to share with you how easy to create an object detection Model with YOLO using data sets from RoboFlow, the main but of our Code is to create a model that can distinguish the players from video and generate an output video with the players tagged

first of all, we install the necessary Libraries:

%pip install ultralytics
!pip install torch torchvision -U
!pip3 install opencv-python        

After the installation we download the data set from RoboFlow

!curl -L "https://universe.roboflow.com/ds/sO3gyQBJYk?key=u4DrcatDg6" > roboflow.zip; unzip roboflow.zip; rm roboflow.zip        

this command will extract the dataset folder for our model training

Next, we execute this code :

from ultralytics import YOLO
import torch
import os

# Load the YOLOv8 model
model = YOLO('yolov8n.pt')

# TRAINING
if __name__ == '__main__':
    results = model.train(data="data.yaml", epochs=50, patience=5)
        

which will create a Yolov8 model and execute the training function with 3 parameters

• data="data.yaml": This argument specifies the dataset that the model will use for training.
• epochs=50: This specifies the number of complete passes through the entire dataset that the training process will execute.
• patience=5: This means the training process will stop if there is no improvement in the model's validation performance for 5 consecutive epochs.

After running the training function we will get a runs folder that contains all the trained models as a result :

The only Model we need is the "best.pt" so that we can use it in our code as follows:

import cv2
from ultralytics import YOLO

# Load the YOLOv8 model
# model = YOLO('yolov8n.pt')          ### Pre-trained weights

model = YOLO('runs/detect/train5/weights/best.pt')          ### weights from trained model

# Open the video file
video_path = r"est-jsk.mp4"
cap = cv2.VideoCapture(video_path)

# Default resolutions of the frame are obtained.
#The default resolutions are system dependent.
# We convert the resolutions from float to integer.
frame_width = int(cap.get(3))
frame_height = int(cap.get(4))

out = cv2.VideoWriter('ResultVideo.mp4',cv2.VideoWriter_fourcc('M','J','P','G'), 20, (frame_width,frame_height))

# Loop through the video frames
while cap.isOpened():
    # Read a frame from the video
    success, frame = cap.read()

    if success:
        # Run YOLOv8 tracking on the frame, persisting tracks between frames
        results = model.track(frame, persist=True, show=False, tracker="botsort.yaml")

        # Visualize the results on the frame
        annotated_frame = results[0].plot()

        # Write the frame into the file 'ResultVideo.mp4'
        out.write(annotated_frame)

        # Break the loop if 'q' is pressed
        if cv2.waitKey(1) & 0xFF == ord("q"):
            break
    else:
        # Break the loop if the end of the video is reached
        break

# Release the video capture object and close the display window
cap.release()
out.release()