The consumer program is adapted from MotionDSP's more powerful product, Ikena (about $7,700), used by law enforcement authorities to recover details like license plate numbers in low-quality video, analyses the color and position of pixels in frames adjacent to ones with the poor images, says CEO Sean Varah, (right) who adds that the information found in the nearby frames make improvements. Enhanced videos can then be saved or uploaded to sites.

Image-enhancement algorithms are part of a research field called super-resolution, said Sanjay J. Patel, a professor of electrical and computer engineering at the University of Illinois at Urbana-Champaign. “Super-resolution is a class of techniques one can use on ordinary video to make it look better.” Those include the computationally intensive process of seeking extra, compensatory information in nearby frames

“There’s a lot of interest in the technology in research labs at universities and in specialised professional software for video processing,” he said — for instance, in crime labs. “But you don’t see much of it in consumer products.”

MotionDSP’s software, called vReveal, can do some editing — trimming clips or rotating sideways video, for example — but its main function is narrower: to improve appearance by increasing resolution and smoothing out the effects of a bobbing camera.

The software’s job is to fix shaky, noisy home video, said Dr Nikola Bozinovic, VP of engineering at MotionDSP. “It’s pretty much for any standard-definition video you’ve recorded, including anything transferred from VHS.”

The number of people who are shooting video or watching it at home is skyrocketing, said Mary Madden, a senior researcher at the Pew Internet and American Life Project. “The medium has been percolating through the online world at a rate that far surpasses other online activities,” she said. and the market for video enhancement may also expand as video calling and conferencing become more mobile. Skype, for instance, has demonstrated mobile video calling from a desktop computer to a mobile Internet device over WiFi.

“The difficulty with this kind of product is that you have to see it or use it to appreciate it,” said Dr Bozinovic. To address this problem, vReveal is offering a one-month free membership, during which people can compare their before-and-after videos, then decide whether to buy the program.

While vReveal works with Windows XP or Vista  (not with Macs), it will make its enhancements much faster if the machine contains a recent graphics processing card from Nvidia, Dr. Varah said. Nvidia is an investor and a marketing partner with vReveal; a specific list of cards is at vReveal’s Web site. “If you have other graphics cards, your computer will just use its central processing unit,” Dr. Varah said, but the process will be slower.

Inspiration by human eye anatomy and physiology
Researchers involved in the study say that fields of research such as action and object recognition, surveillance, wide-base stereo microscopy and three-dimensional shape reconstruction could benefit a great deal frominnovation, bringing about a new wave of fresh and powerful devices. Details of the new technology will be presented  at the upcoming annual IEEE meeting on computer vision. In charge of the new research were computer scientists Hao Jiang and Stella X. Yu, both from the Boston College.

In earlier versions of similar systems, a video camera mounted on the computer would capture the image, and the computer then browse through millions of pictures, looking for a possible match. The time- and resource-consuming, task made more difficult by moving objects, where shifting orientations and angles continuously modified the search parameters that the search software used, forcing it to start over.

 In the new method, the research team was able to devise a novel solution of linear algorithms, which is able to make the computer's work a lot easier.

“When the human eye searches for an object it looks globally for the rough location, size and orientation of the object. Then it zeros in on the details. Our method behaves in a similar fashion, using a linear approximation to explore the search space globally and quickly; then it works to identify the moving object by frequently updating trust search regions,”  explains Jiang, (right)  a BC assistant professor of computer science.

Unlike other viewing methods, which only have a 50% success rate – consuming enormous amounts of resources during a session – new technology picks up objects  with a 95% accuracy, at only a fraction of the hardware complexity of its predecessors. It relies on mathematical formulas to identify an object,  functioning  a lot faster and more efficiently than comparing data-bank pictures obtained by an outside camera.