News for hackers. Stuff that matters.

A comment on Russia’s new FindFace app identifies strangers in a crowd with 70% accuracy. Parent comment

Re:Russia’s new FindFace app identifies strangers in a crowd... by aub3bhat on Wednesday May 18, 2016 @09:40AM (#11721800)
The Google FaceNet and other similar system typically perform an approximate nearest neighbor search in a ~128 dimensional embedding of faces.

Here is the paper for the reference: http://arxiv.org/pdf/1503.03832v3.pdf

Also 70% accuracy is like really really bad.

Here are results from MegaFace Challenge, which involved identifying face in presence of Million distractors:

http://megaface.cs.washington.edu/results/fgnetresults.html

[ Parent | on HN ]
Re:Russia’s new FindFace app identifies strangers in a crowd... by matheweis on Wednesday May 18, 2016 @10:03AM (#11721991)
That was with the FGNet dataset... Ntech labs (incidentally the algorithm behind FindFace) beat Googlr on the FaceScrub dataset.

http://megaface.cs.washington.edu/results/facescrubresults.h...

[ Parent | on HN ]
Re:Russia’s new FindFace app identifies strangers in a crowd... by pradn on Wednesday May 18, 2016 @02:32PM (#11724477)
The accuracy is not 70%. The percentage of times it is able to match a face to a social media profile, given the person has a profile, is 70%. You can't compare apples-to-apples with the results in Google's FaceNet.
[ Parent | on HN ]