tracking

Submitted by Sek Chai

Scientists at the Autonomous University of Barcelona's Computer Vision Center have developed a prototype system that interprets sign language. It can interpret more than 20 Spanish signs. A video camera is used to record the video sequence, and computer vision software is used to detect face, hand and arm movements. A classification algorithm then corresponds each movement with the appropriate sign. More information is available in a press release.

Marketwatch reports that Lockheed-Martin is acquiring markerless motion capture technology from startup Organic Motion. The STAGE(TM) Military Tracking System (MTS), which was demonstrated at ITEC London this week, will be used for supporting systems for simulations and training.

Submitted by Sek Chai

PrimeSense, an Israeli startup, recently revealed that it will be providing the 3D motion sensor for Microsoft’s Project Natal. The motion-control system would be used in the Xbox 360 to let players control an on-screen avatar using full body gestures. More information is available in a press release.

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.

At the 2009 E3 conference, Microsoft presented a 3-D motion sensing camera for its Xbox 360 video game consoles. The system, code-named Project Natal, is intended to replace control pads and joysticks. It includes a camera and depth sensor to track body gestures and build depth maps. The camera system also includes face and voice recognition to track gamers in the living room.

A day after Microsoft introduced Project Natal, Sony demonstrated a controller with a glowing sphere at the end that a PlayStation Eye camera can track. The controller includes internal motion sensors that can track movements within 7mm accuracy.

The SVEN (Surveillance Video and Entertainment Network) performance art project created music videos automatically from live surveillance camera input. The SVEN software for tracking people in uncontrolled environments is available for download. The software also produces mattes of the people being tracked to support visual effects. The software is in C++ and uses the OpenCV 1.0 library. Real-time performance at 30 frames per second has been achieved at resolution of 640x480 pixels.