Semiconductor IP News and Trends Blog
Qualcomm Recognizes Importance of Gesture IP
Another patent sale makes the news. This time, telecom chip giant Qualcomm is buying gesture-recognition technology from GestureTek. The chip company plans to incorporate gesture recognition technology into its “current and next generation Snapdragon processors.” The target market is smart phones, tablets and other consumer devices.
This acquisition brings two important points to light. First, this gesture recognition technology is tied to a specific processor instead of a unique operating system. To me, this linkage acknowledges the relative newness of the processor-cycle intensive algorithms needed for this technology.
Many intelligent applications, like gesture and face detection-recognition, use multiple-camera systems spread over a given space. These cameras often use CMOS sensor imaging technology to meet low-power constraints of the embedded space. Advance DSP-based image processing is required to condition the signals from the multiple cameras as well as help with the actual gesture recognition calculations.
The second important point of this announcement is that it serves as yet another example of the continuing growth of sensor system-related IP.
As with any evolving technology, questions abound. For example, what specific design algorithms’ are involved? This question might become important in any legal battle between GestureTek and Microsoft (Kinect) over gesture recognition systems.
One a side note, Microsoft’s recent release of their Kinect SDK has spread quickly throughout the world of robotics. Among its other capabilities, the SDK enables development of gesture-facial recognition and skeletal tracking. One cool application of this technology is the “jedibot,’ a foam lightsaber yielding robotic arm created by students at Stanford University. Students Create JediBot
Switching from the design to the manufacturing side, one might wonder if challenges will arise from the fabrication of the MEMS sensor platform used in the CMOS cameras or other sensors instrumentation systems. Perhaps this will be less of an issue with camera (facial) systems as opposed to full-body motion capture (think Avatar) platforms. One example of such a platform is Movea’s MotionPod, a wireless motion sensing system that incorporates a 3-axis accelerometer, a 3-axis gyroscope and a 3-axis magnetometer.
This is an exciting time in the development of gesture recognition and full-body motion systems. As usual, the gaming and movie markets are the first to embrace these user-interface intensive technologies. But Qualcomm’s acquisition, as well as the ongoing success of motion-sensitive mobile devices like Apple’s iPhone and iPad, proves these technologies are headed for the mainstream consumer spaces.