Phil Edwards, who works as a senior video producer for Vox, looked into why AI art has so much trouble accurately reproducing hands.
Producer Phil Edwards igs into why AI art struggles with hands. The challenges range from the same ones that human artists face to those that are a unique result of how AI generative art is created. The road to improving these hands may not be as obvious as you’d think.
Edwards turned to artist Stan Prokopenko, MIT Robotics student Yilun Du, and scientist Roy Shilkrot to learn more about the mechanics of reproducing a hand.
Talking to them helped me figure out three big reasons. Not every reason, but three big reasons. That hands are tough for AI art models. The data size and quality. The way hands act and the low margin for error.
Edwards also talked about how steps are being taken to correct this issue.
And they are working on two things. First, they have the AI look at a ton more pictures which requires more computing power. …The other solution might be to invite more people… into the museum. ….They basically fine tune the model so that it would generate sentences that are convincing to people.