What I like most about announcements like this is that it shows how a new generation of bio-inspired vision insights are slowly working their way into commercial products.
It's easy to forget how astonishingly fast and energy efficient real-time object recognition and tracking is in biological sysems. But once innovators recognize what's possible and start to take the implications to heart, they enter a new competitive domain of building and selling vision-capable products that in time can achieve amazingly good price points. The old mathematically-inspired approach of treating "vision" strictly as a problem in how to process enormous 3D (2D x time) floating point matrices works after a fashion, but only by burning up an amazing number of CPU cycles and more than a little energy. Conversely, if you are simply reading this text right now, you are yourself an existence proof that more efficient approaches to parsing complex visual information are possible.
It seems like image processing is going to be more and more on our minds moving forward. I keep seeing these amazing concepts that are only barely able to function with current tech. Can't wait to see what near future improvements bring to this area.
I believe Ambric was an acquisition by Nethera as well and provided the hardware architecture used for the stream processing algorithms Nethera developed. I believe it allows video data to flow thru the system with portions of the algorithms executed at each step, simplifying and speeding overall processing.
Would be good to have someone else dig into the details however, my data is a few years old now.
What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.