Deep Learning: Computer Vision

What is the course about?

How do self-driving cars recognize pedestrians? How do smartphones unlock with a glance? How do robots “see” the world? This course explores the answers through deep learning and computer vision—the field that enables machines to interpret and respond to visual information.

In this hands-on course, students will learn how to design intelligent systems capable of analysing images, detecting objects, and recognizing patterns in visual data. From powering facial recognition and medical diagnostics to enabling augmented reality and surveillance systems, computer vision lies at the heart of many modern AI applications. Using Python and powerful libraries such as TensorFlow, Keras, and PyTorch, learners will build their own vision-based models and gain insight into how algorithms learn to “see” and make decisions based on what they observe.

What will we cover?

The course begins with the foundations of image processing and neural networks, leading into the core concepts of convolutional neural networks (CNNs)—the backbone of modern computer vision. Students will explore image classification, multi-label prediction, transfer learning, and data augmentation. Advanced topics include object detection using models such as YOLO and SSD, facial recognition, visual search, and explainability through tools like Grad-CAM. The course also addresses real-world case studies, including visual quality control in manufacturing, anomaly detection in surveillance, and vision systems for robotics and healthcare.

What will I achieve?
By the end of this course you should be able to...

By the end of the course, students will be able to:

• Recognise the principles of how machines interpret visual input using deep learning
• Build, train, and evaluate convolutional neural networks for image classification
• Apply techniques such as transfer learning to fine-tune powerful pre-trained models
• Develop real-time object detection and facial recognition systems
• Preprocess, manage, and augment image datasets for optimal model performance
• Apply computer vision solutions to real-world challenges in fields such as security, automation, and healthcare

What level is the course and do I need any particular skills?

This course requires proficiency in Python (students should have completed our introduction and intermediate Python courses or have equivalent knowledge or experience of the topics covered in these courses) and completion of the Introduction to Machine Learning course. While previous deep learning experience is advantageous, it is not essential, as concepts will be thoroughly explained and reinforced through structured, progressive exercises. Ideal for students, developers, researchers, and professionals passionate about applying AI to visual challenges.

How will I be taught, and will there be any work outside the class?

The course is structured around hands-on, project-driven learning. Each topic is introduced through a combination of short lectures, real-world case studies, and live coding sessions. Students will work on guided lab exercises and progressively develop their own computer vision projects. While most of the work is completed in class, learners are encouraged to explore additional datasets or experiment with advanced model tuning outside of sessions to deepen their understanding and creativity.

Are there any other costs? Is there anything I need to bring?

There are no additional costs for this course. All necessary materials and software, including access to OpenCV, will be provided.

When I've finished, what course can I do next?

Please click here to view our Programming and Maths courses

Disclaimer: Use of Third-Party Software
This course might require you to either use your own personal account or create an account for the purposes of this course. City Lit cannot accept any responsibility for any failings of the third party or provide technical support. Whilst using the software you will be responsible for abiding by the providers terms and conditions and maintaining your own work.