Invariant

InVariant and Equavariant

How to invariant detection, but still get the exact position

How to detect general voice but separate people

How to detect face, but separate people

Common mode detection

And differential mode detection