Pose estimation is a computer vision technique that determines the posture of people, or the position and orientation of objects, in images or video. Typically a model detects key points such as shoulders, elbows and knees and joins them into a skeleton. It is used, for example, in fitness apps that check exercises, in workplace ergonomics analysis, in sports analytics and in robotics.
Computer vision is a field of AI in which computers interpret images and video. Typical tasks include detecting and locating objects, reading text in images and measuring components. The technology is used, for example, in quality control in manufacturing, in driver assistance systems, in logistics and in medical imaging.
Image recognition is a field of computer vision in which software identifies what an image shows – objects, people, text, logos or scenes. It is usually based on neural networks trained on large numbers of labelled images. Applications range from quality control in manufacturing and sorting product photos to supporting medical imaging. With images of people, data protection and potential bias require particular attention.
Depth estimation determines from images how far objects or individual pixels are from the camera. This can be done with two cameras (stereo), with depth sensors or, using neural networks, even from a single photo (monocular depth estimation). It is needed in robotics, driver assistance, augmented reality and 3D models of rooms.
A convolutional neural network (CNN) is a neural network architecture particularly suited to image data. It scans the input with small filters, detecting simple features such as edges in the first layers and more complex shapes in deeper ones. CNNs shaped image recognition for many years and are used, for example, to spot defects in product photos or in medical image analysis.
