Human perception of the outside world is formed by perceiving images which is one of the most important perception channels and giving them meaning by analyzing. All techniques aiming to create on computers visual perception and understanding capability of humans fall into the field of computer vision and cognitive learning. Scientists working in the field of computer vision have played the biggest role in the advancement of deep learning which is the most mentioned sub-field of artificial intelligence today.
Computer vision aims to make sense of two-dimensional, three-dimensional or higher-dimensional all kinds of visual digital data, especially with smart algorithms. In solving these problems, the field of computer vision uses the development and implementation of both mathematical and computational theory based techniques such as geometry, linear algebra, probability and statistics theory, differential equations, graph theory and especially in recent years machine learning and deep learning. In addition to standard camera images, medical images, satellite images, and computer modeling of three-dimensional objects and scenes are of interest to computer vision.
Visual data detection and interpretation constitute the most important algorithm steps in many applications, from autonomous systems such as autonomous vehicles, robots, and drones to security and biometric verification areas. The outputs obtained are fed as inputs to the decision support systems in the next step, completing the artificial intelligence systems.
Prof. Dr. Gözde Ünal, one of our faculty members who has been working in this field for more than 20 years, has more than 30 patents in American and European patent offices. In Google Scholar keyword searches, it ranks first in Turkey and 421st in the world with deep learning, and 1320th in the world with computer vision. She is still the director of the ITU Vision Lab, where research studies are carried out actively in the field of Computer Vision under the Faculty of Computer and Informatics Engineering at ITU.