Archive for the ‘Computer vision’ Category

Grassroots Mapping

Wednesday, May 26th, 2010

This is a balloon photography blog. The authors try to help to citizens with low resources to do mapping.

Grassroots Mapping.

Machine vision and image processing: Oil spill fifty times worse than BP predicts

Sunday, May 23rd, 2010

Machine vision and image processing: Oil spill fifty times worse than BP predicts.

John Nack on Adobe: Video: New from Adobe Labs, Content-Aware Fill in Photoshop

Friday, March 26th, 2010

John Nack on Adobe: Video: New from Adobe Labs, Content-Aware Fill in Photoshop.

Free Gimp tutorial PDFs | TuxRadar Linux

Tuesday, March 2nd, 2010

Free Gimp tutorial PDFs | TuxRadar Linux.

Recognizr: An Augmented Identity App – GOOD Blog – GOOD

Thursday, February 25th, 2010

Recognizr: An Augmented Identity App

via Recognizr: An Augmented Identity App – GOOD Blog – GOOD.

The MathWorks releases new software update

Wednesday, February 17th, 2010

Matlab and Image Processing Toolbox now offer more multi-threaded implementations of functions, and Statistics Toolbox adds parallel computing support for several functions.

via The MathWorks releases new software update.

Red Bull Illume - Image Quest 2010

Wednesday, February 17th, 2010

Beat the Deadline! Enter before Feb 28

Red Bull Illume - Image Quest 2010.

Using Lasers to Zap Mosquitoes – Bits Blog – NYTimes.com

Sunday, February 14th, 2010

Nathan Myhrvold, Microsoft’s former chief technology officer, thinks so. His company, Intellectual Ventures, has assembled commonly available technology — parts used in printers, digital cameras and projectors — to make rapid lasers to shoot down mosquitoes in mid-flight. If bed nets are the low-tech solution to combat the deadly disease — caused by a parasite transmitted when certain mosquitoes bite people — the laser is a high-tech one.

via Using Lasers to Zap Mosquitoes – Bits Blog – NYTimes.com.

Visual Neuroscience Group’s 18-inch Cube Supercomputer | VizWorld.com

Thursday, February 11th, 2010

“A few months ago, nVidia generously donated 8 of their impressive 9800GX2 graphics cards to our lab, to help us scale-up our efforts. The resulting machine is described below. With peak performance around 4 TFLOPs (4 billion floating point operations per second), this little 18”x18”x18” cube is perhaps one of the world’s most compact and inexpensive supercomputers.”

via Visual Neuroscience Group’s 18-inch Cube Supercomputer | VizWorld.com.

VideoTrace: Rapid interactive scene modelling from video

Monday, January 14th, 2008

The idea is very interesting. Use computer vision techniques for 3D reconstruction but using human intervencion.
I recomend going to http://www.acvt.com.au/research/videotrace/ and see the demonstration video. Results are very impressive.

What i think it lies behing is a visual localization and mapping algoritm that reconstructs the points given by the human intervencion. when a point is lost it relies on the operator again to recover from the error.

There are more technical details on thte paper on the web