[sagebird]

I’m interested in presentation detection and enhancement in general. For example, if a person is standing in front of a mostly static chalkboard for an hour, there is a ton of information there about the writing on the board. The frames could be fed into a handwriting detector and perhaps vectorized, and the video format should understand this and transmit a full resolution chalkboard on frame zero of the transmission, and not send any board info until there is an actual modification.