There is a perceptron the ID all object. A second generative NN that transform the object position into the next frame, a prediction from past data. Self training happens by comparing two frame of data a , eye, percepton activates when dead center over target and then moves to second frame. The eye searches for target and when eye percepton is over target again then eye percepton activates again. The eye perceptron change in position as a input into the generator nn for prediction. No real math just gradient of decent and remembering.
A few pixels are enough to restore the video in virtual reality: https://www.youtube.com/watch?v=p1GBFpNvI6M&feature=youtu.be Neural Network - Lens: https://www.youtube.com/watch?v=92NqMLSaZKI ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/Te623675bfcbd1b22-M08fb4a1b605cb9ff83ad043c Delivery options: https://agi.topicbox.com/groups/agi/subscription
