Re:Show HN: Neural Image Compression Demo
by tonic_section on Saturday September 12, 2020 @03:31PM
(#24454870)
Yeah, high frequency detail such as facial features for faraway figures or text tend to get washed out after compression - this is probably due to a couple reasons: 1) The training dataset contains relatively few pictures including text, 2) high-frequency detail is too expensive to encode and the model learns to forgo encoding this in favor of more 'important' features such as shapes, colors, etc.
> [4]: Sample reconstructions
The text in the reconstructed image in the third row looks different, the word phonomat is quite garbled, information looks a bit funny.