Re:Images that fool computer vision raise security concerns
by kevin_thibedeau on Monday March 23, 2015 @12:41PM
(#9251791)
It's not excellent research. All they did is create a naive NN without any feature recognition and noted that it has a high rate of false positives when presented with random imagery. Duh.
[ Parent
| on HN ]
Re:Images that fool computer vision raise security concerns
by woah on Monday March 23, 2015 @12:44PM
(#9251814)
Really? From the article:
"They tested with two widely used DNN systems that have been trained on massive image databases."
Is this wrong?
Re:Images that fool computer vision raise security concerns
by kevin_thibedeau on Monday March 23, 2015 @01:00PM
(#9251924)
GIGO. Just because they used a "Deep" architecture doesn't mean it was suitable for the task.
Re:Images that fool computer vision raise security concerns
by jessaustin on Monday March 23, 2015 @01:06PM
(#9251961)
Does your criticism extend to the whole enterprise of facial recognition, or did this research somehow fail to include the secret sauce that makes it work?