Researchers demonstrate a better way for computers to 'see' (w/ Video)

Dec 02, 2009

Taking inspiration from genetic screening techniques, researchers from Harvard and MIT have demonstrated a way to build better artificial visual systems with the help of low-cost, high-performance gaming hardware.

The neural processing involved in visually recognizing even the simplest object in a natural environment is profound—and profoundly difficult to mimic. Neuroscientists have made broad advances in understanding the visual system, but much of the inner workings of biologically-based systems remain a mystery.

Using Graphics Processing Units (GPUs), the same technology video game designers use to render life-like graphics, researchers are now making progress faster than ever before. A new study, co-led by David Cox, Principal Investigator of the Visual Group at the Rowland Institute at Harvard, and Nicolas Pinto, a Ph.D. Candidate in James DiCarlo's laboratory at the McGovern Institute for Brain Research and the Department of Brain and Cognitive Sciences at MIT, was published in the November 26th issue of PLoS Computational Biology.

"Reverse engineering a biological visual system—a system with hundreds of millions of processing units—and building an artificial system that works the same way is a daunting task," says Cox. "It is not enough to simply assemble together a huge amount of computing power. We have to figure out how to put all the parts together so that they can do what our brains can do."

"While studying the brain has yielded critical information about how the brain is wired, we currently don't have enough information to build a computer system that works like the brain does," adds Pinto. "Even if we take all of the clues that we have available from experimental neuroscience, there is still an enormous range of possible models for us to explore."

To tackle this problem, the team drew inspiration from screening techniques in molecular biology, where a multitude of candidate organisms or compounds are screened in parallel to find those that have a particular property of interest. Rather than building a single model and seeing how well it could recognize visual objects, the team constructed thousands of candidate models, and screened for those that performed best on an object recognition task.

The resulting models outperformed a crop of state-of-the-art computer vision systems across a range of test sets, more accurately identifying a range of objects on random natural backgrounds with variation in position, scale, and rotation.

Using ordinary computer processing units, the effort would have required either years of time or millions of dollars of computing hardware. Instead, by harnessing modern graphics hardware, the analysis was done in just one week, and at a small fraction of the cost.

"GPUs are a real game-changer for scientific computing. We made a powerful parallel computing system from cheap, readily available off-the-shelf components, delivering over hundred-fold speed-ups relative to conventional methods," says Pinto. "With this expanded computational power, we can discover new vision models that traditional methods miss."

This high-throughput approach could be applied to other areas of computer vision, such as face identification, object tracking, pedestrian detection for automotive applications, and gesture and action recognition. Moreover, as scientists understand better what components make a good artificial vision system, they can use these hints when studying real brains to understand them better as well.

"Reverse and forward engineering the brain is a virtuous cycle. The more we learn about one, the more we can learn about the other," says Cox. "Tightly coupling experimental neuroscience and engineering holds the promise to greatly accelerate both fields."

Source: Harvard University (news : web)

Explore further: Coping with floods—of water and data

add to favorites email to friend print save as pdf

Related Stories

Professor working on practical cloaking device (w/ Video)

Aug 20, 2009

(PhysOrg.com) -- A physicist at the University of St Andrews hopes to make major advances in the 'tantalising' field of invisibility in the next two years. Professor Ulf Leonhardt, who cites the Invisible Woman and Harry ...

Recommended for you

Coping with floods—of water and data

Dec 19, 2014

Halloween 2013 brought real terror to an Austin, Texas, neighborhood, when a flash flood killed four residents and damaged roughly 1,200 homes. Following torrential rains, Onion Creek swept over its banks and inundated the ...

Cloud computing helps make sense of cloud forests

Dec 17, 2014

The forests that surround Campos do Jordao are among the foggiest places on Earth. With a canopy shrouded in mist much of time, these are the renowned cloud forests of the Brazilian state of São Paulo. It is here that researchers ...

User comments : 4

Adjust slider to filter visible comments by rank

Display comments: newest first

rincewind
5 / 5 (1) Dec 02, 2009
Using ordinary computer processing units, the effort would have required either years of time or millions of dollars of computing hardware. Instead, by harnessing modern graphics hardware, the analysis was done in just one week, and at a small fraction of the cost.


As the authors say, "reverse and forward engineering the brain" is a virtuous cycle. This methodology is poised to explode in the next two decades.

Massively exciting article.
SincerelyTwo
not rated yet Dec 02, 2009
HP Labs, which developed the memristos, is currently developing a artificial neural net system exploiting their properties.

Google for more info, awesome stuff ahead.
Birger
not rated yet Dec 03, 2009
Hmmm....David Jones (aka "Daedalus") in the magazine "Nature" predicted that as CCTV cameras become ubiquitous, and pattern recognition software advance to the level that any face can be recognised in real time, the anonymity of the cities may go away and be replaced by the kind of social control that existed (and still exist) in rural villages. Everyone will be a participant of the ultimate reality show....
Birger
not rated yet Dec 03, 2009
As a consequence of "robotic vision" becoming viable with smaller computers, the future could bring small autonomous munitions that are indifferent to the usual decoys. Vehicles and even humans could be spotted instantly, and since small munitions can escape detection until activated an attack across open terrain could get as costly as the battle of the Somme. Main battle tanks can not have thick armor all over, and since munitions with "vision" can home in on the weak spots, the domination of the tank will be seriously challenged.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.