The disciplines of Computer Vision and Deep Learning make it possible to train algorithms so that machines can learn patterns from images. This learning process can be accelerated through the use of computational power of Graphics Processing Units (GPUs).
In a business context, the following application can be implemented:
Object Detection
A technique that makes it possible to locate the position of objects (people, cars, beverages, etc.) within an image or video frame. To do so, the algorithm must be trained with a previously labeled set of images.
Face Recognition
A technique that allows the detection and indentification of people based on the physiological characteristics of their faces in an image or video. To do so, the algorithm needs to analyze a determined set of images for each person.