FaceNet: A Unified Embedding for Face Recognition and Clustering · HackerPapers discussions

FaceNet: A Unified Embedding for Face Recognition and Clustering — mentioned by bko

Sun, 07 Jul 2019 22:33:05 GMT

A single picture is surprisingly effective in recognizing that person in another photo. I've been playing around with a python wrapper on dlib's facial recognition [0] and it's astonishing how it can identify people from just one photo. It's not like other machine learning methods that include training a classifier or even fine tuning a classifier. It works by encoding the persons face to a 128 dimension vector and then using that for comparison. The accuracy from most methods is between 99.2% - 99.8%, but the problem is that the training samples are too easy and controlled. It's sensitive to…

Paper and discussion history. A mention is not an endorsement.

FaceNet: A Unified Embedding for Face Recognition and Clustering — mentioned by paulsutter

Sat, 11 Aug 2018 14:53:04 GMT

The authors didn't try very hard to solve the problem (although it's still slightly interesting). Many networks can detect if two never-before-seen objects are the same or different. Simple example is Koch's Siamese Neural Networks[1]. More useful example is Facenet[2], which can tell if two faces are the same or different, even if never trained on these specific faces. Facenet learns an n-dimensional embedding and similarity is estimated by euclidean distance. These are cool papers by the way. The Omniglot dataset[3] used in Koch is a sort of an "inverse MNIST". Omniglot has 20 examples each…

Paper and discussion history. A mention is not an endorsement.

FaceNet: A Unified Embedding for Face Recognition and Clustering — mentioned by halflings

Fri, 16 Jun 2017 19:33:48 GMT

They would probably use FaceNet[0] then, if they only want to detect faces, as that should give better results. [0] https://arxiv.org/abs/1503.03832

Paper and discussion history. A mention is not an endorsement.

FaceNet: A Unified Embedding for Face Recognition and Clustering — mentioned by nl

Wed, 17 May 2017 17:15:29 GMT

A single NN can predict more than one class of object. The ImageNet competition has 20,000 classes. There's also image segmentation as another poster has pointed to. In the case of FB face tagging, they'd have learn an embedding space for faces, and when a new image comes in they'd place it in the embedding space along with all the person's connections and find the nearest neighbors. See https://arxiv.org/abs/1503.03832 or the implementation https://cmusatyalab.github.io/openface/

Paper and discussion history. A mention is not an endorsement.

FaceNet: A Unified Embedding for Face Recognition and Clustering — mentioned by aub3bhat

Wed, 18 May 2016 13:40:11 GMT

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

Paper and discussion history. A mention is not an endorsement.

FaceNet: A Unified Embedding for Face Recognition and Clustering — mentioned by gwern

Wed, 13 Apr 2016 20:53:48 GMT

I am not sure this works or that success rates like claimed are possible: > In 70 percent of cases, he was quite easily able to identify the people he had photographed. VKontakte has 280m+ accounts and is the second most popular website in Russia (https://en.wikipedia.org/wiki/VK_%28social_networking%29); the Russian population is ~144m. I can't find any decent estimate of how many Russians use VK, but it seems to be at least as dominate in Russia as Facebook in the USA, which is to say, ~71% (http://www.pewinternet.org/fact-sheets/social-networking-fac...). So from non-VK usage alone, he…

Paper and discussion history. A mention is not an endorsement.

FaceNet: A Unified Embedding for Face Recognition and Clustering — HN submission

Mon, 23 Mar 2015 02:43:42 GMT

FaceNet: A Unified Embedding for Face Recognition and Clustering

2 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

FaceNet: A Unified Embedding for Face Recognition and Clustering — HN submission

Fri, 13 Mar 2015 19:18:34 GMT

Google FaceNet: A Unified Embedding for Face Recognition and Clustering

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.