CogniVue announces real-time computer vision processors
EE Times reports that Canadian company CogniVue has announced a new family of processors for real-time video analysis. The multicore CV220X processors are intended for a broad range of applications and include capabilities for detection and recognition of both objects and people. Automotive-quality versions are available, and the company sees automotive applications as one leading market for these chips.
Related Computer Vision Central news:
- Login or register to post comments
- Printer-friendly
VisLab autonomous vehicle to drive 13,000 km from Italy to China
Submitted by Massimo Bertozzi
VisLab Intercontinental Autonomous Challenge
The World Expo 2010 that will be held in Shanghai, China, is the third most relevant worldwide event after the FIFA World Cup and the Olympic Games, 70 millions of visitors from all over the world are expected to visit the Expo. This year’s theme is “better cities, better life”; issues related to sustainable mobility are indeed central to the Expo, which will be a display of new ideas developed worldwide in this field.
VisLab is going to participate to the Expo with a project, partially funded by ERC - European Research Council- that will show that it is possible, although in a prototype version, to move goods between two continents with non-polluting vehicles powered by green energy and with virtually no human intervention. Some goods will be packed in Rome, some collected throughout the trip, and finally taken to Shanghai on vehicles with no driver for the first time in history.
VisLab will be responsible for vehicle technology and automation iissues, while Overland will be responsible for logistics and media events.
The aim is to demonstrate, through an extensive and impressive test, that the current technology is mature enough for the deployment of non-polluting and no-oil based autonomous vehicles in real conditions. Moreover the Municipality of Rome, an active player in this project, is planning to exploit these vehicles downtown to deliver goods to shops, collect trash, and arrange sustainable mobility in the last mile.
Two electric vehicles will perform a 13,000 km trip mainly powered by solar energy, with no driver; two backup vehicles will be part of the trip as well. As a support, 4 Overland trucks will follow the expedition to provide a mechanic shop, storage, and accommodation; finally two additional trucks will be used for media coverage and will be equipped for live satellite broadcasting.
The first vehicle will drive autonomously in selected sections of the trip and will conduct experimental tests on sensing, decision, and control subsystems, and will continuously collect data. Although limited, human interventions will be needed to define the route and intervene in critical situations.
The second vehicle will automatically follow the route defined by the preceding vehicle, requiring no human intervention (100% autonomous). This will be regarded as a readily exploitable vehicle, able to move on predefined routes; at the end of the trip, its technology will be transferred to a set of vehicles to move in the inner part of Rome in the close future.
During the trip, demonstrations will be performed in specific hot spots; autonomous vehicles will follow given routes, negotiating traffic, avoiding obstacles, and stopping when required. The first demonstration was held in Rome at the EUR district on October 29, 2009, when Rome's Major, Gianni Alemanno, officially presented this challenge. More information on the presentation can be found in the official press release or slides.
Enhanced rover vision automatically seeks locations to photograph
Stephen Pugh, a researcher from Aberystwyth University, is developing algorithms for robots to quickly discover locations of interest and choose targets for pictures without human intervention. He is working on panoramic cameras in robotic planetary rovers within the ExoMars Programme – a project led by the European Space Agency with NASA. More information is available in a WalesOnline article.
- Login or register to post comments
- Printer-friendly
Fixstars releases Linux OS for CUDA
Fixstars Corporation, a Tokyo-based multicore software provider, has released Yellow Dog Enterprise Linux (YDEL) for CUDA. The operating system is aimed at NVIDIA GPUs to accelerate graphics and GPGPU applications (such as financial services, image processing). According to the HPCWire article, until recently, Fixstars focused mainly on Power CPUs and the IBM Cell Broadband Engine. The YDEL release represents the "first commercial Linux distribution for GPU computing".
- Login or register to post comments
- Printer-friendly
Sony unveils new motion-controller
Sony Corp has unveiled the Move controller for its PlayStation gaming console. The controller can be used with the PS3 Eye gaming webcam to translate user motions into game actions. The controller is shaped like a TV remote with a colorful ball at one end. More information is available in a press release.
Computer Vision Central has been following this story -- see previous articles:
- Login or register to post comments
- Printer-friendly
Point Grey reveals ultra compact Flea3 camera
Point Grey recently introduced their Flea3 ultra compact cameras that measure only 29 x 29 x 30 mm. The camera incorporates a FireWire 800-Mbps interface, with mono and color versions of an image sensor from Sony. According to the product website, Flea3 offers new features such as opto-isolated GPIO for industrial triggering and strobe output, non-volatile flash memory for user data storage, and an on-camera frame buffer.
- Login or register to post comments
- Printer-friendly
Google integrates user photos in Street View
Google has introduced a new way to view user photos on its Street View application. According to a blog posted by Daniel Filip and Daniel Cotting, (Computer Vision Team, Google Zürich), "user photos allow you to view locations from entirely new perspectives". A new interface in the upper right corner allows viewing of a variety of user-submitted photos from Picasa, Panoramio, and Flickr.
- Login or register to post comments
- Printer-friendly
Motion tracking and facial animation combined in partnership of two companies
Xsens Technologies, which sells an inertial motion capture suit, and Image Metrics, which provides facial animation services based on computer vision methods, have partnered to provide a complete body and face animation solution. The combined solution allows actors full freedom of motion. The technology is already in use by a special effects studio, Double Negative. More information is available in a press release.
- Login or register to post comments
- Printer-friendly
Computer Vision to be discussed at SXSW festival panel
South by Southwest (SXSW), originally a music festival, has grown into a collection of festivals and conferences covering music, film, and emergent interactive technologies. This year, there will be a panel titled What If Your Phone Had Five Senses? featuring Ted Power and Nicholas Jitkoff of Google, Matt Biddulph of Doppir, and Ben Averch of Microvision. Among other areas, this panel will touch on augmented reality, speech recognition, and computer vision. The panel discussion will take place March 14th in Austin, Texas, USA.
- Login or register to post comments
- Printer-friendly
Extended Deadline: 2010 Embedded Computer Vision Workshop
The 2010 Embedded Computer Vision Workshop deadline for submissions has been extended to March 19, 2010.
See the ECVW call for papers for complete information on the workshop, including on-line submission instructions.
Disclosure: ECVW 2010 is hosted on this website.
- Login or register to post comments
- Printer-friendly
