Video facial recognition and facial feature points detection continues to evolve using event-based cameras. Creating a hybrid of frame based and event based camera technologies is proving to create high accuracy lip reading potential with video cameras.
As the technology progresses, facial feature points will be processed through deep learning to determine the orientation of the face so that the lips and their movement can be read and analyzed.
This technology should provide some interesting applications (such as speech to text for the deaf) but will certainly lead to amazing legal challenges.
Stay tuned.
Commentaires