The effects are pretty cool, but I can't help but wonder how much better it would be if done with a bit of intention and knowledge of what these algorithms are actually doing.
"We found that while the raw data makes no music, the waveforms share patterns"
Its not that surprising, since music is frequency based while pictures are spatially based. I expect you could get something that sounded like music from a picture if you did some Fourier-transform filtering before with different interval domains (different sizes of 'chunks' being converted to frequencies).
These days there are a number of audio synthesizers which do basically what you said, such as the Beepmap plugin that comes stock with FL Studio.
Among the first to use the technique artistically, Aphex Twin famously drew his own face in the spectrum of his track "Equation" (https://www.youtube.com/watch?v=wSYAZnQmffg - skip to 5:20 for the face).
"We found that while the raw data makes no music, the waveforms share patterns"
Its not that surprising, since music is frequency based while pictures are spatially based. I expect you could get something that sounded like music from a picture if you did some Fourier-transform filtering before with different interval domains (different sizes of 'chunks' being converted to frequencies).