SEGMENT – 3D scene understanding in two glances

Project Description:

In project “SEGMENT”, Professor Thomas Wallis and his team are investigating what is known as visual segmentation. This is when the human brain attempts to recognise which parts of a scene shown, for instance, in a photo belong to which objects. The researchers want to find out how people combine information from active 3D vision, and how they can learn this combination from visual experiences. To this end, they want to develop a new display device and experimental methods, and to simulate inputs into the visual system from early development to adulthood. The plan is to use groundbreaking new technologies based on advanced computer graphics and machine learning.

Computer experiments with artificial neural networks are also planned to help in understanding segmentation learning. The scientists at Darmstadt want to systematically restrict or manipulate various factors such as optics or eye movements. At the end, the learned behaviour of the artificial networks will be compared with the behaviour of adults during segmentation in the active exploration of 3D scenes. “We want to use similarities and differences to better understand a fundamental mystery in perception: how the mind recognises the meaning of scenes,” explained Professor Wallis.

Contact:

Thomas Wallis has been a Professor of Perception at the Institute of Psychology and Center for Cognitive Science at TU Darmstadt since 2021. His research focuses on visual perception in humans and machines, machine learning and cognitive modelling, and applications in visual perception.

Australian-born Wallis received his doctorate from the University of Queensland in Australia in 2010. He then conducted research at Schepens Eye Research Institute and Harvard Medical School in the USA as well as at the Centre for Integrative Neurosciences and the Department of Computer Science at the Eberhard Karl University of Tübingen. Thomas Wallis was a researcher at Amazon Deutschland from 2019 until 2021.