UCLA computer scientists, statisticians, mathematicians and psychologists are devising new ways of looking at the world in the new W. M. Keck Foundation Lab in Vision and Image Sciences at UCLA.

Through a $670,000 grant from the Keck Foundation, UCLA will establish a central location for researchers in the UCLA Collective on Vision and Image Sciences, formed in 2002.

“There are a large number of researchers at UCLA interested in vision, and the Keck Lab will help bring them together,” said computer science professor Stefano Soatto. “The UCLA Vision Collective is very successful already, but having a shared physical location will help spur new collaborations.”

The Keck Foundation grant will equip a state-of-the-art facility at UCLA for measuring visual environment, creating computer vision models, and testing vision with virtual reality. The new lab will have high-end scanners, motion capture cameras and systems, a geodesic light dome, and a computer cluster for vision models. The set-up will allow researchers to identify the properties of images that support recognition by machines and humans, offering insights into the brain and artificial intelligence.

Led by statistics professor Alan Yuille, the UCLA researchers hope to pave the way for advances in understanding human vision and creating machine vision by studying the visual world. The interdisciplinary team is developing computer visual systems that register and interpret images in much the same way as human eyes and minds.

“Researchers working with speech signals know that words are made up of basic building blocks known as phonemes, and that makes it a lot easier for them to build speech recognition systems that work,” explained Yuille. “The visual system is more complicated because images have not yet been systemically studied and modeled. Part of our aim is to understand the complexity of the visual environment in order to build vision systems that work, but we also want to understand how biological systems, which have evolved to deal with complexity, function.”

“The best visual system we have is the human eye and brain,” noted Soatto. “But it’s incredibly complicated – you use half of your brain to process visual information. We want to build an artificial visual system that operates as effectively, and the Keck Lab will offer us many new exciting opportunities to address this challenge.”

While people are able to easily infer information about an environment by looking at it, it is much more difficult for machines to gather information from unknown or unstructured environments. Soatto will lead the effort to measure environments in ways that machines can process and understand.

Potential applications for advanced computer vision systems include improved security systems and automation, and assistance for the visually impaired.

The Keck Lab also will benefit students at UCLA who are interested in vision. Undergraduate and graduate students will have an opportunity to apply theories from their interdisciplinary courses to projects in the lab, working with their colleagues from other disciplines.

Based in Los Angeles, the W. M. Keck Foundation was established in 1954 by the late W. M. Keck, founder of the Superior Oil Company. The Foundation’s grant making is focused primarily on pioneering efforts in the areas of medical research, science and engineering.