(Page 2 of 2)
In a recent demonstration at Intel's Santa Clara campus, Nestares ran the algorithm on a dual-core Intel machine to add resolution to a webcam (click here to see a video of a demonstration). The algorithm was able to add four times the resolution to captured video, with no noticeable delay. Nestares says that the algorithm is written to scale to the number of cores that it runs on: the more cores available, the more the processing gets divided up. "Different regions [of the video] are assigned to different cores," says Nestares. Using this method, it would take about 150 cores to convert DVD video into HD DVD in real time, he says.
There's an economic driver to be able to add resolution to video, says William Freeman, a professor of electrical engineering and computer science at MIT. More people are buying HD TV sets, and there's still a lot of content that hasn't caught up with the resolution that those sets can provide. "There's a real need ... People are going to want to look at high-definition content on their new HD TV sets," Freeman says.
At this point, however, Intel doesn't have any plans to make the algorithm commercially available. But Bautista says that the team has talked with product groups at Intel about the possibility.
A new website provides an innovative tool that expands and contracts images while minimizing distortion.
Other super-resolution solutions
Nothing new. Just google a bit and you'll see that this super resolution technology already exists as an Adobe After Affects and Premiere Pro plug-in and also as a stand-alone Windows application. It actively uses multiple cores too and works much faster (320x240 -> 640x480 with 20 fps on a quad core).
http://www.thedeemon.com/super_resolution/
http://www.thedeemon.com/VideoEnhancer/
Found something from CNBC about super-resolution in the film industry. Looks like all the indie filmmakers are doing backflips now.
http://www.cnbc.com/id/23954891/
The company's called Topaz Labs. www.topazlabs.com . They make some bold claims.
Manufacturing in the United States is in trouble. That's bad news not just for the country's economy but for the future of innovation.
This document is part of the “How-To Guide for Most Common Measurements” centralized resource portal. This tutorial provides a detailed guide for measurement and device considerations to take temperature measurements using thermocouples. Get an introduction to thermocouples, which are inexpensive sensing devices widely used with PC-based data acquisition systems. Also review some specific thermocouple examples and learn how thermocouples work and ways to integrate them into a data acquisition measurement system.
View full PDF >
Listen to story > 
Our list of the 50 most innovative companies, including the following:
On Twitter
Become a Fan on Facebook
Subscribe to the Feed
inboulder
10 Comments
Nice article, poor presentation
The video on this site of the upscaling in question is so small as to be useless. This is especially ironic considering the content of this article. Did Tech Rev contemplate this issue before publishing?
Reply
Mr.Doerr
1 Comment
Re: Nice article, poor presentation
I'm sorry that you did not see the difference in the video, which by the way was clearly visible. The idea was to show you an example, not to market the product.
Reply
3danim8
1 Comment
Re: Nice article, poor presentation
Not only was the video in the left and right panels indistinguishable, the text that described the video panels was also in error. It stated: "The original video, on the left, was captured with a standard USB 2.0 webcam with an input of 160 by 120 pixels. The output on the left is 320 by 240 pixels."
Should the description read: "output on the right..."
Reply
brunascle
65 Comments
Re: Nice article, poor presentation
i agree. at this scale it's hard to make a judgement. i can see a difference between the two, but i wouldnt call the one on the right _better_, just different. the one on the left is blurry, the one on the right is blocky. and they're both equally legible, at least IMO.
and judging the result by the resolution you're able to achieve is totally meaningless. i can put a video into whatever resolution i want simply by filling in random pixels, or upscaling. the only way to judge it is to either compare it to an original taken at the same resolution, or by human perception.
Reply
GaryB
119 Comments
Re: Nice article, poor presentation
While I agree that say an 85 year old would see no difference in these videos, those of us who are younger and retain some visual acuity can quite clearly read the text better on the right than on the left.
I'd like to see this work on non-text, say a sports clip with near/play far/crowd differences. Or better yet, a youtube clip which have just awful resolution and terrible compression artifacts. But, as is, it's pretty cool work.
Reply
mem11363
1 Comment
Re: Nice article, poor presentation
I believe the original video steadily cycles from clear to blurry to clear. This is consistent with what I have experienced with webcams when the amount of motion in the input stream exceeds their processing capability. The "enhanced" video remains clear. So this demonstration tells me that the algorithm is successful at determining the clarity of each frame it processes and uses the better ones - ignores or puts a lower weight on the worse ones. But that seems like a different goal than taking a good quality video at resolution X (like a high quality dvd) and converting it to a high quality video at say double or quadruple the initial resolution.
Reply