FaceNet

FaceNet is a facial recognition system developed by Florian Schroff, Dmitry Kalenichenko and James Philbina, a group of researchers affiliated with Google. The system was first presented at the 2015 IEEE Conference on Computer Vision and Pattern Recognition.^[1] The system uses a deep convolutional neural network to learn a mapping (also called an embedding) from a set of face images to a 128-dimensional Euclidean space, and assesses the similarity between faces based on the square of the Euclidean distance between the images' corresponding normalized vectors in the 128-dimensional Euclidean space. The system uses the triplet loss function as its cost function and introduced a new online triplet mining method. The system achieved an accuracy of 99.63%, which is the highest score to date on the Labeled Faces in the Wild dataset using the unrestricted with labeled outside data protocol.^[2]

The structure of FaceNet is represented schematically in Figure 1.

For training, researchers used input batches of about 1800 images. For each identity represented in the input batches, there were 40 similar images of that identity and several randomly selected images of other identities. These batches were fed to a deep convolutional neural network, which was trained using stochastic gradient descent with standard backpropagation and the Adaptive Gradient Optimizer (AdaGrad) algorithm. The learning rate was initially set at 0.05, which was later lowered while finalizing the model.

The researchers used two types of architectures, which they called NN1 and NN2, and explored their trade-offs. The practical differences between the models lie in the difference of parameters and FLOPS. The details of the NN1 model are presented in the table below.

FaceNet introduced a novel loss function called "triplet loss". This function is defined using triplets of training images of the form ${\displaystyle (A,P,N)}$. In each triplet, ${\displaystyle A}$ (called an "anchor image") denotes a reference image of a particular identity, ${\displaystyle P}$ (called a "positive image") denotes another image of the same identity in image ${\displaystyle A}$, and ${\displaystyle N}$ (called a "negative image") denotes a randomly selected image of an identity different from the identity in image ${\displaystyle A}$ and ${\displaystyle P}$.

Let ${\displaystyle x}$ be some image and let ${\displaystyle f(x)}$ be the embedding of ${\displaystyle x}$ in the 128-dimensional Euclidean space. It shall be assumed that the L2-norm of ${\displaystyle f(x)}$ is unity (the L2 norm of a vector ${\displaystyle X}$ in a finite dimensional Euclidean space is denoted by ${\displaystyle \Vert X\Vert }$.) We assemble ${\displaystyle m}$ triplets of images from the training dataset. The goal of training here is to ensure that, after learning, the following condition (called the "triplet constraint") is satisfied by all triplets ${\displaystyle (A^{(i)},P^{(i)},N^{(i)})}$ in the training data set:

${\displaystyle \Vert f(A^{(i)})-f(P^{(i)})\Vert _{2}^{2}+\alpha <\Vert f(A^{(i)})-f(N^{(i)})\Vert _{2}^{2}}$

The variable ${\displaystyle \alpha }$ is a hyperparameter called the margin, and its value must be set manually. Its value has been set as 0.2.

Thus, the full form of the function to be minimized is the following function, which is officially called the triplet loss function:

${\displaystyle L=\sum _{i=1}^{m}\max {\Big (}\Vert f(A^{(i)})-f(P^{(i)})\Vert _{2}^{2}-\Vert f(A^{(i)})-f(N^{(i)})\Vert _{2}^{2}+\alpha ,0{\Big )}}$

In general, the number of triplets of the form ${\displaystyle (A^{(i)},P^{(i)},N^{(i)})}$ is very large. To make computations faster, the Google researchers considered only those triplets which violate the triplet constraint. For this, for a given anchor image ${\displaystyle A^{(i)}}$ they chose that positive image ${\displaystyle P^{(i)}}$ for which ${\displaystyle \Vert f(A^{(i)})-f(P^{(i)})\Vert _{2}^{2}}$ is maximum (such a positive image was called a "hard positive image") and that negative image ${\displaystyle N^{(i)}}$ for which ${\displaystyle \Vert f(A^{(i)})-f(N^{(i)})\Vert _{2}^{2}}$ is minimum (such a positive image was called a "hard negative image"). since using the whole training data set to determine the hard positive and hard negative images was computationally expensive and infeasible, the researchers experimented with several methods for selecting the triplets.

• Generate triplets offline computing the minimum and maximum on a subset of the data.
• Generate triplets online by selecting the hard positive/negative examples from within a mini-batch.

On the widely used Labeled Faces in the Wild (LFW) dataset, the FaceNet system achieved an accuracy of 99.63% which is the highest score on LFW in the unrestricted with labeled outside data protocol.^[2] On YouTube Faces DB the system achieved an accuracy of 95.12%.^[1]

• DeepFace
• FindFace

• Rajesh Gopakumar; Karunagar A; Kotegar, M.; Vishal Anand (September 2023). "A Quantitative Study on the FaceNet System": in Proceedings of ICACCP 2023. Singapore: Springer Nature. pp. 211–222. ISBN 9789819942848.
• Ivan William; De Rosal Ignatius Moses Setiadi; Eko Hari Rachmawanto; Heru Agus Santoso; Christy Atika Sari (2019). "Face Recognition using FaceNet (Survey, Performance Test, and Comparison)" in Proceedings of Fourth International Conference on Informatics and Computing. IEEE Xplore. doi:10.1109/ICIC47613.2019.8985786. Retrieved 6 October 2023.
• For a discussion on the vulnerabilities of Facenet-based face recognition algorithms in applications to the Deepfake videos: Pavel Korshunov; Sébastien Marcel (2022). "The Threat of Deepfakes to Computer and Human Visions" in: Handbook of Digital Face Manipulation and Detection From DeepFakes to Morphing Attacks (PDF). Springer. pp. 97–114. ISBN 978-3-030-87664-7. Retrieved 5 October 2023.
• For a discussion on applying FaceNet for verifying faces in Android: Vasco Correia Veloso (January 2022). Hands-On Artificial Intelligence for Android. BPB Publications. ISBN 9789355510242. Amazon