Vision

This Computer Vision Central article reports the new NVIDIA CUDA 4.0 toolkit includes a computer vision library. From the article: "NVIDIA announced the latest version of the NVIDIA® CUDA® Toolkit for developing parallel applications using NVIDIA GPUs.

The NVIDIA CUDA 4.0 Toolkit was designed to make parallel programming easier, and enable more developers to port their applications to GPUs. This has resulted in three main features:

NVIDIA GPUDirect™ 2.0 Technology -- Offers support for peer-to-peer communication among GPUs within a single server or workstation. This enables easier and faster multi-GPU programming and application performance.

This Computer Vision Central article looks at an interactive art installations made by Petronio A. Bendito that use computer vision software to analyze the motion and appearance of visitor-participants. The nstallation called Action//Musique plays music according to the visitor's movements and determines colors in a wall projection according to the visitor's clothing. Additional information can be found in a JConline article.

This iFixit article provides a look inside the Microsoft Kinect device. Here's the related video.

This press release reports Canesta has signed a definitive agreement to have its products, technology, intellectual property, customer contracts, and other resources acquired by the Microsoft Corporation. From the article: "Canesta is a leader in 3-D sensing technologywhich is critical to making Natural User Interfaces (NUI) possible.

According to Jim Spare, Canesta president and CEO, “This is very exciting news for the industry. There is little question that within the next decade we will see natural user interfaces become common for input across all devices. With Microsoft’s breadth of scope from enterprise to consumer products, market presence, and commitment to NUI, we are confident that our technology will see wide adoption across many applications that embody the full potential of the technology.”

This Computer Vision Central article looks at the EU-funded MIAUCE project which studies how a computer can extract information from seeing human behavior. From the article: "The goal is to develop a form of ambient intelligence in which the computer's multimodal interface enables people to "interact with their environment". There are three applications being developed: a surveillance system to detect situations that affect safety, a monitor to observe how customers shop, and an interactive Web TV that enables users to select what they want to see based on which part of the screen they are viewing.

This ScienceDaily article reports a new method devised by computer scientists at Carnegie Mellon University enables computers to gain a deeper understanding of an image by reasoning about the physical constraints of the scene. From the article: "

In much the same way that a child might use a set of toy building blocks to assemble something that looks like a building depicted on the cover of the toy set, the computer would analyze an outdoor scene by using virtual blocks to build a three-dimensional approximation of the image that makes sense based on volume and mass.

This Physorg.com article is looking at an object recognition system, developed by MIT researchers, which breaks images into ever smaller parts. From the article: "Object recognition is one of the core topics in computer vision research: After all, a computer that can see isn't much use if it has no idea what it's looking at. Researchers at MIT, working with colleagues at the University of California, Los Angeles, have developed new techniques that should make object recognition systems much easier to build and should enable them use computer memory more efficiently.

This Wired article talks about a flight simulator built to understand how flies can process images much faster than humans. From the article: "Ever wonder how an insect with such a tiny brain can thwart your attempts to catch it nearly every time? Apparently scientists do, too.

To find out how the common blowfly manages to process visual images more than four times faster than humans, researchers have built the bug a flight simulator.

From this Science Daily article : "Vast quantities of data are transferred in real time from telescopes around the world to a supercomputer in the Netherlands, where European researchers combine the information to create high-resolution images of distant objects in space.

By pointing up to 16 radio telescopes from six continents at one source in space and combining the observation signals from the telescopes via a high-speed network, European astronomers have created a ‘virtual telescope’ that delivers better resolution than any single telescope on earth. The high-speed network also makes it easier for astronomers to react to so-called Targets of Opportunity – transient events such as supernova explosions and gamma-ray bursts in space.

This ZDNet article reports "a team of scientists at the Fraunhofer Institute for Photonic Microsystems IPMS in Dresden, Germany, is working on a device which incorporates eye tracking to influence the content presented to the viewer. Without having to use any other devices to enter instructions, the wearer can display new content, scroll through a menu or shift picture elements simply by moving her eyes or fixing on certain points in the image.

“We want to make the eyeglasses bidirectional and interactive so that new areas of application can be opened up,” says Dr. Michael Scholles, business unit manager at IPMS.