Computer vision is the AI field that gives computers the ability to understand images and video. CNNs (Convolutional Neural Networks) learn images hierarchically: early layers detect edges, middle layers shapes, final layers objects. AlexNet's victory at the ImageNet competition (2012) revolutionised this field.
Core tasks: Image classification (which class does this belong to?), Object detection (YOLO — identifying multiple objects in an image), Semantic segmentation (assigning a class to every pixel), Pose estimation (determining human body position). Practical applications: medical image analysis, autonomous vehicles, security cameras, and quality control in manufacturing.
Our courses on this topic