Generating 2D and 3D Master Faces for Dictionary Attacks with a Network-Assisted Latent Space Evolution

  title={Generating 2D and 3D Master Faces for Dictionary Attacks with a Network-Assisted Latent Space Evolution},
  author={Tomer Friedlander and Ron Shmelkin and Lior Wolf},
—A master face is a face image that passes face-based identity authentication for a high percentage of the population. These faces can be used to impersonate, with a high probability of success, any user, without having access to any user information. We optimize these faces for 2D and 3D face verification models, by using an evolutionary algorithm in the latent embedding space of the StyleGAN face generator. For 2D face verification, multiple evolutionary strategies are compared, and we propose… 



Generating Master Faces for Dictionary Attacks with a Network-Assisted Latent Space Evolution

The results presented demonstrate that it is possible to obtain a high coverage of the LFW identities with less than 10 master faces, for three leading deep face recognition systems.

FaceNet: A unified embedding for face recognition and clustering

A system that directly learns a mapping from face images to a compact Euclidean space where distances directly correspond to a measure offace similarity, and achieves state-of-the-art face recognition performance using only 128-bytes perface.

Learning from Millions of 3D Scans for Large-Scale 3D Face Recognition

  • S. Z. GilaniA. Mian
  • Computer Science
    2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
  • 2018
A method for generating a large corpus of labeled 3D face identities and their multiple instances for training and a protocol for merging the most challenging existing 3D datasets for testing and the efficacy of this method for the open world face recognition problem is shown.

Generating Master Faces for Use in Performing Wolf Attacks on Face Recognition Systems

It is demonstrated that wolf (generic) faces, which are called “master faces,” can also compromise face recognition systems and that the master face concept can be generalized in some cases.

Deep 3D face identification

It is shown that transfer learning from a CNN trained on 2D face images can effectively work for 3D face recognition by fine-tuning the CNN with an extremely small number of 3D facial scans.

Learning Effective Intrinsic Features to Boost 3D-Based Face Recognition

A novel representation, called intrinsic features, is presented to encode local 3D shapes and describes complementary non-relational features to provide an intrinsic representation of faces to produce consistently better results than existing methods.

Master Face Attacks on Face Recognition Systems

An extensive study on latent variable evolution (LVE) is reported, hypothesize that master faces originate in dense areas in the embedding spaces of face recognition systems, and demonstrates that the existence of master faces poses an actual threat.

DeepFakes: a New Threat to Face Recognition? Assessment and Detection

This paper presents the first publicly available set of Deepfake videos generated from videos of VidTIMIT database, and demonstrates that GAN-generated Deep fake videos are challenging for both face recognition systems and existing detection methods.

A 3D Face Model for Pose and Illumination Invariant Face Recognition

This paper publishes a generative 3D shape and texture model, the Basel Face Model (BFM), and demonstrates its application to several face recognition task and publishes a set of detailed recognition and reconstruction results on standard databases to allow complete algorithm comparisons.

On the (Limited) Generalization of MasterFace Attacks and Its Relation to the Capacity of Face Representations

This work comprehensively analyse the generalizability of MasterFace attacks in empirical and theoretical investigations using six state-of-the-art face recognition models, cross-dataset and cross-model evaluation protocols, and utilizing testing datasets of significantly higher size and variance.