Hi Ui,What I think would be interesting to see would be Akidas ability to learn autonomously. There used to be a video using DVS data from inilabs, where Akida would learn to recognise cars on a highway and start to "count" them
I wonder if this could be applied to things in the blood to identify new things that shouldn't be there
I think you've hit the nail on the head.
Event cameras are good at spotting events, ie, detecting moving objects. They can do this without Akida. Where Akida comes in is in identifying/classifying the events (or groups of events), and in learning from the events.
So if the job is just counting unidentified objects, there is no need for Akida, but if the job requires identification of objects, Akida is your man.
Last edited: