Machine vision research is an important part of the development of artificial intelligence: Prof. Janne Heikkilä from CMVS was interviewed by Academy of Finland
“In image recognition, machine vision outperformed the human eye years ago. Now, the challenge is to take advantage of three-dimensional information in observing the environment", says machine vision researcher Janne Heikkilä.
Self-guided vehicles and other robots are coming. Robotic cars must be able to "see" people, obstacles and other objects while moving, so machine vision is of great importance in robotics. In a three-dimensional world, machine vision applications must be able to identify objects and evaluate distances, and position themselves on the map by means of image-based positioning.
"For a long time, machine vision meant analysis of two-dimensional images, mainly image recognition. However, the world is three-dimensional, and machine vision needs to be able to detect and understand the three-dimensional structure of the view, " explains Janne Heikkilä, Computer Science Professor at Center for Machine Vision and Signal Analysis (CMVS) at the University of Oulu.
“In image recognition, machine vision outperformed the human eye years ago. Now, the challenge is to take advantage of three-dimensional information in observing the environment."
A research project funded by the Academy of Finland and lead by Professor Heikkilä develop methods for image-based identification, modeling and positioning using a 3D machine vision. The results of the research are mainly used in applications of augmented and virtual reality, and for example in creating virtual copies of objects.
Link to the story in Academy of Finland: http://www.aka.fi/fi/akatemia/media/Ajankohtaiset-uutiset/2018/konenakotutkimus-tarkea-osa-tekoalyn-kehittymista/
Last updated: 14.8.2018