YOLOv8 Object Detection with DeepSORT Tracking(ID + Trails)

Resumen esencial

El código está funcionando correctamente. Se arreglaron todos los elementos deprecados del código, todos los archivos están en carpeta, y para usar el código con un video en específico, se aplica la siguiente línea en la terminal:

• Clonar proyecto

git clone https://github.com/SamuelEPradoT/YOLOv8-DeepSORT-Object-Tracking.git

• Ir a la carpeta de proyecto

cd YOLOv8-DeepSORT-Object-Tracking

• Instalar dependencias

pip install -e '.[dev]'
pip install torch

• Identificar carpeta del código en cuestión (ahí se dejan los videos input).

cd ultralytics/yolo/v8/detect

• Aplicar código (Video: video_traffic_2.mp4).

python predict.py model=yolov8l.pt source="video_traffic_2.mp4" show=True

El output queda en la carpeta: YOLOv8-DeepSORT-Object-Tracking\runs\detect

Por hacer

• Modificar código para que la línea sea proporcional al tamaño de los frames del video
• Implementar FLASK para que el output sea redirigido a la API.

Google Colab File Link (A Single Click Solution)

The google colab file link for yolov8 object detection and tracking is provided below, you can check the implementation in Google Colab, and its a single click implementation, you just need to select the Run Time as GPU, and click on Run All.

Google Colab File

Object Detection and Tracking (ID + Trails) using YOLOv8 on Custom Data

Google Colab File Link (A Single Click Solution)

Google Colab File

YOLOv8 Segmentation with DeepSORT Object Tracking

Github Repo Link

Steps to run Code

• Clone the repository (Proyecto clonado)

git clone https://github.com/SamuelEPradoT/YOLOv8-DeepSORT-Object-Tracking.git

• Goto the cloned folder.

cd YOLOv8-DeepSORT-Object-Tracking

• Install the dependecies

pip install -e '.[dev]'

• Setting the Directory.

cd ultralytics/yolo/v8/detect

• Downloading the DeepSORT Files From The Google Drive (Ya descargado)

https://drive.google.com/drive/folders/1kna8eWGrSfzaR6DtNJ8_GchGgPMv3VC8?usp=sharing

• After downloading the DeepSORT Zip file from the drive, unzip it go into the subfolders and place the deep_sort_pytorch folder into the yolo/v8/detect folder (Ya descargado)

• Downloading a Sample Video from the Google Drive (Ya descargado)

gdown "https://drive.google.com/uc?id=1rjBn8Fl1E_9d0EMVtL24S9aNQOJAveR5&confirm=t"

• Run the code with mentioned command below.

• For yolov8 object detection + Tracking

python predict.py model=yolov8l.pt source="test3.mp4" show=True

• For yolov8 object detection + Tracking + Vehicle Counting (Esto falta por hacer)
• Download the updated predict.py file from the Google Drive and place it into ultralytics/yolo/v8/detect folder
• Google Drive Link

https://drive.google.com/drive/folders/1awlzTGHBBAn_2pKCkLFADMd1EN_rJETW?usp=sharing

• For yolov8 object detection + Tracking + Vehicle Counting

python predict.py model=yolov8l.pt source="test3.mp4" show=True

RESULTS

Vehicles Detection, Tracking and Counting

Vehicles Detection, Tracking and Counting

Watch the Complete Step by Step Explanation

• Video Tutorial Link YouTube Link