An Intellyx Brain Candy Brief
The business and civil development applications of AI-based image and video recognition for smart cities, human-centered design and massive hyperconverged video networks will reach the mass market, once we cross certain price, power and form factor barriers.
Anything that reduces the cost, energy consumption, and size footprint of the enabling architecture while increasing performance can only help usher in this new wave of visual computing.
Gyrfalcon Technology Inc. (or GTI) offers a specialized range of low-cost, low-power AI accelerator chips, advanced processing cards, aggregated ASIC hardware, AI development toolkits and plug-in testing dongles for developers and designers to build a visual computing cortex into devices, computers and hybrid cloud servers.
Their MPE (Matrix Processing Engine) statistics demonstrate a step-change leap over previous CPU and GPU-based approaches to AI interpretation of 2D image streams. The details of the accelerated TOPS/W (Tera operations per second / by watt) efficiency and performance stats for CNN (convoluted neural network) processing going on in these chips may translate into more malleable miniaturized device designs, 5G edge networks, centralized servers and cloud-based AI applications that tilt the market in favor of widespread adoption.
© 2020 Intellyx. At the time of writing, Gyrfalcon Technology is not an Intellyx customer. None of the other vendors mentioned here are Intellyx clients. Want to see more BrainCandy? Subscribe today. If you are a vendor seeking coverage from Intellyx, please contact us at PR@intellyx.com.
