Computer vision may not be as good as thought

Jan 25, 2008
Computer vision may not be as good as thought
The human brain easily recognizes that these cars are all the same object, but the variations in the car's size, orientation and position are a challenge for computer-vision algorithms. Image / Nicolas Pinto

For years, scientists have been trying to teach computers how to see like humans, and recent research has seemed to show computers making progress in recognizing visual objects. A new MIT study, however, cautions that this apparent success may be misleading because the tests being used are inadvertently stacked in favor of computers.

Computer vision is important for applications ranging from “intelligent” cars to visual prosthetics for the blind. Recent computational models show apparently impressive progress, boasting 60-percent success rates in classifying natural photographic image sets. These include the widely used Caltech101 database, intended to test computer vision algorithms against the variety of images seen in the real world.

However, James DiCarlo, a neuroscientist in the McGovern Institute for Brain Research at MIT, graduate student Nicolas Pinto and David Cox of the Rowland Harvard Institute argue that these image sets have design flaws that enable computers to succeed where they would fail with more authentically varied images. For example, photographers tend to center objects in a frame and to prefer certain views and contexts. The visual system, by contrast, encounters objects in a much broader range of conditions.

“The ease with which we recognize visual objects belies the computational difficulty of this feat,” explains DiCarlo, senior author of the study in the online Jan. 25 PLoS Computational Biology. “The core challenge is image variation. Any given object can cast innumerable images onto the retina depending on its position, distance, orientation, lighting and background.”

The team exposed the flaws in current tests of computer object recognition by using a simple “toy” computer model inspired by the earliest steps in the brain's visual pathway. Artificial neurons with properties resembling those in the brain's primary visual cortex analyze each point in the image and capture low-level information about the position and orientation of line boundaries. The model lacks the more sophisticated analysis that happens in later stages of visual processing to extract information about higher-level features of the visual scene such as shapes, surfaces or spaces between objects.

The researchers intended this model as a straw man, expecting it to fail as a way to establish a baseline. When they tested it on the Caltech101 images, however, the model did surprisingly well, with performance similar or better than five state-of-the-art object-recognition systems.

How could that be? “We suspected that the supposedly natural images in current computer vision tests do not really engage the central problem of variability, and that our intuitions about what makes objects hard or easy to recognize are incorrect,” Pinto explains.

To test this idea, the authors designed a more carefully controlled test. Using just two categories-planes and cars-they introduced variations in position, size and orientation that better reflect the range of variation in the real world.

“With only two types of objects to distinguish, this test should have been easier for the 'toy' computer model, but it proved harder,” Cox says. The team's conclusion: “Our model did well on the Caltech101 image set not because it is a good model but because the 'natural' images fail to adequately capture real-world variability.”

As a result, the researchers argue for revamping the current standards and images used by the computer-vision community to compare models and measure progress. Before computers can approach the performance of the human brain, they say, scientists must better understand why the task of object recognition is so difficult and the brain's abilities are so impressive.

Source: Massachusetts Institute of Technology

Explore further: Australia, Malaysia, Indonesia to trial new jet tracking system

add to favorites email to friend print save as pdf

Related Stories

Florentine basilica gets high-tech physical

Feb 26, 2015

Late last year, two University of California, San Diego students set out for Florence, Italy, to diagnose a patient that had no prior medical record, couldn't be poked or prodded in any way, and hadn't been ...

Lenovo stops Superfish preloads and issues advisory

Feb 21, 2015

Lenovo has seen calmer weeks. News sites in droves rang chimes and sirens over an adware program on some Lenovo models escalating to concerns about the potential risk of a Man in the Middle threat. Lenovo ...

Virtual-twin plan could support surgery for soldiers

Feb 16, 2015

A Saturday story in the MIRROR.CO.UK, the online edition of the Daily Mirror, carried a headline that made readers look once, twice, three times: "'Virtual twins' could save lives as doctors prepare to te ...

Video of SmartEyeglass Attach! from Sony is released

Feb 14, 2015

Sony continues to seed interest in its smart eyewear concept. Sony recently released a video showing the SmartEyeglass Attach! in action, sent to YouTube via the Sony's Xperia Development team. The video ...

Recommended for you

Phone firms and the quest for the 5G Holy Grail

37 minutes ago

Lightning-quick downloads, driverless cars and remote surgery: telecom firms are racing to develop a new generation of "5G" mobile networks that could start to change the world in five years.

Fujitsu shows iris recognition system that unlocks phones

47 minutes ago

In the bid to come up with authentication solutions beyond passwords, fingerprint authentication from Qualcomm is making news, and so is Fujitsu's iris recognition, yet another potential authentication tech ...

Supreme Court allows challenge to Colorado Internet tax law

2 hours ago

A unanimous Supreme Court ruled Tuesday that federal courts can hear a dispute over Colorado's Internet tax law. One justice suggested it was time to reconsider the ban on state collection of sales taxes from companies outside ...

New incubator network to help clean-energy entrepreneurs

3 hours ago

The Energy Department's National Renewable Energy Laboratory (NREL) and the Electric Power Research Institute (EPRI) have launched the Clean Energy Incubator Network. The program, funded by the Energy Department, aims to ...

User comments : 0

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.