This isn’t really a problem that’s going to go away on solely image based recognition, the contours of a face with dark skin can be less prominent in an image, if there is diminished lighting it becomes even more of an issue, so the identifying features the camera and software could pick up are lost and image manipulation is limited in how much it can possibly try to accurately bring out again.
It’s also basically why just the past week there was a story about how ‚Juggalo‘ makeup stopped a show venue’s facial id system from being able to track the crowds at an Insane Clown Posse gig.
Systems with depth sensors aren’t quite as limited.
Curious_USA_Human on
I read the post title as „fecal recognition“ and wondered which poor bastard got that job!
DVXC on
What a shock. The thing that everyone has been saying is a problem turns out to be a problem.
Maybe next they can tackle the inherent facism enabled by this tech too. I won’t hold my breath.
the_real_deal_feel on
It’s actually Police Service. Official vocab guidelines state “force” is too aggressive.
redpandafire on
This is getting stupid. It’s a very well studied bias of models and they were warned up front. What is it going to take for people to read studies and not just “I’m smarter than science” everything these days?
Leave A Reply
Du musst angemeldet sein, um einen Kommentar abzugeben.
5 Kommentare
This isn’t really a problem that’s going to go away on solely image based recognition, the contours of a face with dark skin can be less prominent in an image, if there is diminished lighting it becomes even more of an issue, so the identifying features the camera and software could pick up are lost and image manipulation is limited in how much it can possibly try to accurately bring out again.
It’s also basically why just the past week there was a story about how ‚Juggalo‘ makeup stopped a show venue’s facial id system from being able to track the crowds at an Insane Clown Posse gig.
Systems with depth sensors aren’t quite as limited.
I read the post title as „fecal recognition“ and wondered which poor bastard got that job!
What a shock. The thing that everyone has been saying is a problem turns out to be a problem.
Maybe next they can tackle the inherent facism enabled by this tech too. I won’t hold my breath.
It’s actually Police Service. Official vocab guidelines state “force” is too aggressive.
This is getting stupid. It’s a very well studied bias of models and they were warned up front. What is it going to take for people to read studies and not just “I’m smarter than science” everything these days?