Neuromorphic computing

And the goal was to build the first neuromorphic chip designed to perform arbitrary computations with networks of spiking neurons.

And write NeuroMorph.

This is a synthesis tool that will automatically configure Brainstorm to perform the desired computations.

And this is based on theory de-

And so you have something like this.

You have your application specifications.

In Nengo, you generate an NEF network description, and then NeuroMorph will then map that onto your chip, and you deploy your chip in whatever your embedded application is you're interested in.

And so this is intro to how we do this.

So instead of using an entire, you know, like your phone camera, you whip it open, you wave your hand in front of it, it kind of gets blurry.

That's pretty much how all cameras work, with the exception of something called neuromorphics.

Neuromorphics only read the delta.

So we're doing something pretty interesting where we are able to detect, track, and maintain custody of things that would typically be blurred, like your hand in front of the camera.

And where's the problem set for that right now? And it's those group ones, right? We have things like radar, we have RF, we have acoustics, all sorts of different ways of to mitigate threats of these other drones.

This audio is presented by Hacker Noon, where anyone can learn anything about any technology.

Neuromorphic vision versus traditional still frame cameras: Which one wins the surveillance race? By Vishwa Vimukthi Gamudu with Thaje.

Surveillance is not a new word.

It has existed in human civilization since ancient times.

Modern systems have evolved from analog closed-circuit television, CCTV, to digital IP cameras with AI analytics and remote access.

But most of that stack still relies heavily on manual review and on one foundational assumption: that the camera produces a stream of full image frames.

That assumption is exactly what neuromorphic vision breaks.

Traditional cameras.

Traditional chips are like lamps that are constantly burning energy, illuminating a room even when no one is in it.

Neuromorphic chips are like motion sensor lights.

They only turn on when needed.

They only turn on and draw power when there is a physical event to react to.