Udemy
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
Computer Vision: OpenCV, Machine Learning and YOLO
New
Rating: 5.0 out of 5(1 rating)
5 students

Computer Vision: OpenCV, Machine Learning and YOLO

Build image classifiers, YOLO detectors, object trackers and pose apps with scikit-learn, TensorFlow and PyTorch.
Last updated 11/2026
English

What you'll learn

  • Process images and video with Python, NumPy and OpenCV: filtering, edge detection, segmentation and image transformations.
  • Train and evaluate image classifiers with scikit-learn, TensorFlow/Keras and PyTorch, using metrics and error analysis.
  • Build CNNs, prepare datasets, and apply augmentation and transfer learning to improve image classification results.
  • Use YOLO for detection, tracking, segmentation and pose estimation; compare YOLO with DETR and RT-DETR.
  • Build traffic counters, parking monitors, fitness repetition trackers and document scanners with downloadable Python code.
  • Export models to ONNX, investigate CPU inference performance, and run computer vision notebooks and apps locally.

Course content

13 sections • 107 lectures • 10h 24m total length
  • Welcome6:22
  • Before You Begin: Feedback, Support and Responsible Use1:12
  • Development Environment - python3:08
  • Development Environment - JupytherLab6:52
  • Python Basics5:15
  • Python Basics - practice13:49
  • What is ML and CV13:11
  • Libraries: NumPy, OpenCV, PIL5:43
  • Libraries: NumPy, OpenCV, PIL - practice5:08
  • Image Processing Basics5:48
  • Image Processing Basics - practice5:07
  • Quick Check: Introduction

Requirements

  • Basic Python knowledge is helpful. A refresher covers variables, loops, functions and the libraries used in the course.
  • No prior machine learning, deep learning or computer vision experience is required.
  • A computer that can run Python and Jupyter. The demonstrated setup and .bat installers use Windows.
  • Internet access for the initial downloads and installation of Python packages.
  • A dedicated GPU is optional. Most demonstrations run on CPU; larger training jobs take longer.

Description

Build practical computer vision applications with Python, OpenCV, machine learning and YOLO. Follow theory lessons and hands-on demonstrations as you learn to process images, train and evaluate models, detect and track objects, and turn model predictions into working applications.

What you will learn

  • Refresh essential Python skills and work with notebooks, NumPy, Pillow and OpenCV.

  • Load and transform images; explore filtering, edges, segmentation and contours.

  • Build and evaluate image classifiers with scikit-learn, including feature extraction, train/test splits and error analysis.

  • Understand neural networks and convolutional neural networks (CNNs), then explore practical examples with TensorFlow/Keras and PyTorch.

  • Work with image datasets, data augmentation, transfer learning and the course's ARTSCI Shapes-10 examples.

  • Use YOLO for object detection, custom-data training, tracking, instance segmentation, classification and human pose estimation.

  • Explore oriented bounding boxes and text-prompted object detection with YOLO-World, compare YOLO with DETR and RT-DETR, and examine model export and CPU inference.

Practical applications

Bring the techniques together in a Parking Lot Monitor, Traffic Counter, Fitness Tracker and Document Scanner. See how detections, tracking IDs, geometry and application logic combine to solve a task, and explore Streamlit interfaces in selected examples.

What is included

The course combines theory, practical code demonstrations, knowledge-check quizzes and downloadable practice packages. Depending on the lesson, packages include notebooks or scripts, dependency instructions and the assets or models needed for the exercise. Run and inspect the examples locally, then adapt them to explore your own ideas. Code, pretrained models and sample assets retain the licenses stated in their packages.

Who this course is for

For learners who want a structured introduction to computer vision and practical Python applications. Basic Python knowledge is helpful, and a refresher is included; no prior machine learning or computer vision experience is required.

Environment and requirements

The demonstrated environment and batch installers use Windows. Most demonstrations can run on a CPU; a GPU is optional and can reduce training time. An internet connection may be needed to install dependencies or obtain resources.

Use of AI in this course

This course contains the use of artificial intelligence. AI-assisted narration and selected AI-generated visual examples support the explanations and practical demonstrations.

Who this course is for:

  • Python developers and students seeking a structured introduction to computer vision, machine learning and deep learning.
  • Learners who want to build and adapt image classification, detection, tracking and pose projects using downloadable code.