A look at the future of GPUs in vision systems

In a new video, Jon Peddie explains how GPU technology and the bi-directional use of massive pixel count will play a major role in the development of embedded vision systems.

Earlier this month the Embedded Vision Alliance invited Dr. Jon Peddie (president of Jon Peddie Research, publisher of GraphicSpeak) to speak at their monthly meeting. Jon explained how graphics processing unit (GPU) technology is reshaping not only graphics but all industries where massive parallel computing power can be of benefit.

Dr. Jon Peddie speaks to the Embedded Vision Alliance. (Source: EVA)
Dr. Jon Peddie speaks to the Embedded Vision Alliance. (Source: EVA)

As Jon explains in his presentation (video links below), GPU technology is still relatively new, having arrived on the scene in 2002. Most people who understand the word “pixel” think only of points of light on the computer screen, but as Jon explains, pixels are becoming a new bi-directional technology. Sensors import real-world data which is then processed by the GPU as pixels, interpreting the data for the required purpose. It’s a lot to think about, which is why you should view the videos and let Jon explain it.

The video below is a four-minute introduction to Jon’s presentation; to watch the entire 52-minute presentation, visit the Embedded Vision Alliance hosting page: