Cutting-edge technologies from the field of artificial intelligence (AI), in particular deep learning, have revolutionized the field of sports science. Computer vision approaches play a pivotal role in converting visual data (e.g., video recordings of competitions or training sessions) into actionable insights, delivering valuable information to researchers, coaches, analysts, and athletes. The applications of computer vision in sports are diverse and range from player detection, tracking of movements and action recognition to the extraction of key performance indicators (KPIs) and tactical patterns via position data estimation. Hence, keeping pace with the latest developments in deep learning is crucial in order to understand the evolving landscape of computer vision in sports analytics. In this chapter, we provide an overview of both foundations and recent research advancements of computer vision in sports science. We present state-of-the-art algorithms and offer practical insights for key applications in sports analytics, including object detection and tracking, action recognition, and sports field registration. Finally, we address existing challenges and discuss potential future research directions such as generative AI models for sports analytics.