5 ms·
very enlightening, and I'm not just saying that. yeah step 3, the implementation is probably all that matters, but in step 1, are you asking the user to do this
by cellis 19y ago
very enlightening, and I'm not just saying that. yeah step 3, the implementation is probably all that matters, but in step 1, are you asking the user to do this?
- tuukkah 19y agoWhat I wanted to express in the form of this tasteless humour was that AI isn't going to be your first problem, even though it could be part of some potential solutions. Modelling the task is the first step. Perhaps I'm wrong but my understanding of the field of AI is that it's trying to get the effects of human intelligence via computation. That is, you'd need to find a way human intelligence could solve the task. What comes to my mind is some kind of physiological modelling and simulation of the facial tissues. But perhaps it would be enough to do something more superficial such as extract features and their statistics from a lot of young-old pairs of photos and apply the same changes to given photos.