From Pixel to Meaning: Introduction to Computer Vision (engl.)

How do pixels become meaningful? This lecture offers an accessible and practical introduction to computer vision and demonstrates how computers extract information from image data. We begin with the basics of digital image representation: pixels, colour models (e.g. RGB) and simple image operations such as filtering and edge detection. Building on this, we introduce classical computer vision methods. We then move on to learning-based methods, in particular convolutional neural networks: their structure, how they work, training and typical architectures. In addition to the theoretical foundations, key applications such as image classification, object recognition and other fields of application (including photography and robotics) will be discussed using concrete examples. There is a particular emphasis on practical application: participants can try their hand at programming (in Python) through short, hands-on exercises and gain an understanding of typical workflows. Prior programming experience is helpful but not essential.

Level 1: for beginners
Requirements: your own laptop
Credit Points (ECTS): —