A team of researchers from The Australian National University (ANU), Data61 and the Australian Centre for Robotic Vision (ACRV) were awarded the “Marr Prize - Honorable Mention” at the biennial IEEE International Conference on Computer Vision 2017 (ICCV’17). The Marr Prize is the most prestigious paper award in the international computer vision community, conferred at its flagship conference.
Dylan Campbell (ANU PhD student), Dr. Lars Petersson (Data61), Dr. Laurent Kneip (ANU/ShanghaiTech), and Hongdong Li (ANU/ACRV) have received this honourable award at the conference held in October, 2017, Venice. Named after David Marr, the Marr Prize is considered one of the top honours for a computer vision researcher.
The award was given for their paper, titled “Globally-Optimal Inlier-set Maximisation for Simultaneous Camera Pose and Feature Correspondence”, which develops a new method for globally-optimal image-based camera localisation.
“This is an exciting achievement, and the technique that we developed in the paper can be applied to solving the problem of accurate autonomous vehicle re-localization in autonomous driving, and is also useful in robot SLAM, as well as in camera tracking for Virtual and Augmented Reality (VR/AR) applications”, said Hongdong.
This is the first time that an Australian team has won such a prestigious computer vision research award since the first International Conference on Computer Vision held in London in 1987.