Face Tracking for Robotic Host with OpenCV and Raspberry Pi
Just a quick update showing Face Tracking is working on top of Face Recognition, and driving the host eyeballs to lock onto any recognized face. This is a very raw setup, and will be modulated by the Language Core (such that when the system is speaking it will use visual accessing cues to mimic real humans - see NeuroLinguistic Programming ["the Other NLP"] "Eye Accessing Cues" for details.) In this manner the system won't just stare at you intimidatingly, but will seem to speak to you in a thoughtful manner as it accesses internal representations, be they of a particular sensory nature (audio/visual) or if they are remembered or constructed.
Currently I only have two functions hanging off of the vision feed - Face Recognition and Object Recognition - but I expect to have several others including text recognition and OCR set up so that the hosts can read printed text from a book, etc., as well as human activity recognition, human emotion estimation, etc.
I also just received a bunch of goodies from SmoothOn to create the next face mask which will be MUCH more detailed and functional thanks to the Stan Winston schools and their video courses.
Thanks for watching, stay healthy, subscribe for more.