This is the seventh epoch of a series with a combined video and image dataset. Might be more coherent than the later epochs; requires testing.
This model aids in generating clips looking down a woman's blouse.