4 ms·
This article covers three methods, all of which just look for alterations of a source image to find a fast match (in fact, that's the paper referenced). It is s
by SavantIdiot 5y ago
This article covers three methods, all of which just look for alterations of a source image to find a fast match (in fact, that's the paper referenced). It is still a "squint to see if it is similar" test. I was under the impression there were more sophisticated methods that looked for types of images, not just altered known images. Am I misunderstanding?
- chipotle_coyote 5y agoApple's proposed system compares against a database of known images. I can't think of a way to "look for types of images" other than trying to do it with machine learning, which strikes me as fraught with incredible fiasco potential. (The compare-to-a-known-database approach has its own issues, including the ones the article talks about, of course.)
- SavantIdiot 5y agoOk, that's what it is seeming like. Since a crypto hash by definition has to generate a huge hamming distance for a small change, everything i've read about perceptual hashes is just the opposite: they should be tolerant enough of a certain amount of difference.