Computer Vision
Given that the AI field is very broad, this group focuses mainly on three distinct areas, namely, image processing (computer vision), machine learning and intelligent agents. Research is conducted in a wide range of application fields using the respective AI algorithms either individually or in some combination. As has become the norm, there is a continued shift to parallel image processing and machine learning using GPUs.
In all of the AI areas supported, research is aimed at solving real-world problems. Example fields for the application of computer vision techniques are biometrics, gesture recognition, surveillance, face recognition, facial expression recognition and object tracking. Members have extensive experience in the application of computer vision to real- world scenarios as former participants in the Integration of Signed and Verbal Communication: South African Sign Language Recognition, Animation and Translation Group.
Intelligent agents are robots that are able to mimic human behaviour in some way. Using machine learning techniques, it is possible to train a robot to perform a human task. Deep learning can be used in both land and aerial robots allowing these to solve tasks that are too difficult or dangerous for humans to solve.
Prof Dane Brown, Prof Karen Bradshaw, James Connan, Josh van Staden and Marc Marais
Last Modified: Thu, 13 Mar 2025 11:35:12 SAST