Computer Vision Research Awarded a Best Paper at International Conference
Congratulations to a team of researchers from School of Engineering and Computer Science from being awarded a Best Paper on their computer vision utilising evolutionary computation work.
, Saud Syed Naqvi
, Will Browne
, Christopher Hollitt
and Mengjie Zhang
were awarded one of 11 best papers (out of 544 submissions ~ 2% awarded BP), which is voted on by peer reviewers and audience members at the presentation at GECCO 2014, Vancouver, Canada, July 2014. Genetic and Evolutionary Computation Conference (GECCO) is one of the most prestigious double-blind peer review conference in Evolutionary Computation. Based on its impact factor, GECCO is 11th in the rankings of 701 international conferences in artificial intelligence, machine learning, robotics, and human-computer interactions.
Their novel contribution is in salient object detection, which is equivalent to identifying the most important object in a photograph. This branch of artificial intelligence is gaining rapid importance due to applications as varied as 'Facebook' image identification to autonomous robotics.
Below is a series of pictures showing the raw image (ASD dataset), then human identified ground truth, followed by two alternative algorithms and the final column showing the state-of-the-art results produced by the developed algorithm.
This work represents a collaboration between the Evolutionary Computational Research Group
and the Vision, Image Computation, and Computer Graphics Group
, which enhances the international reputation. Both groups are currently seeking excellent doctorate students to continue this research direction in computer graphics, vision techniques, evolutionary computation and robotics, where University scholarships are available for suitable candidates.