[Garet]

Yeah, standard APIs tend to drop the ball as soon as lighting or angles get a bit messy, which drove us nuts. For custom face recognition we didn’t want an off-the-shelf API. Went with Data Science UA for a tailored model and the accuracy on our dataset was noticeably better. It actually recognized edge cases properly instead of throwing random false positives, which saved us a ton of headaches during testing.

[Author]

Posts