DeepVisage: Making face recognition simple yet with powerful generalization skills

Hasnat, Abul; Bohné, Julien; Milgram, Jonathan; Gentric, Stéphane; Chen, Liming

doi:10.48550/arxiv.1703.08388

Cited by 13 publications

(14 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The objective is to train a base model F so that its face representations maximizes inter-subject separation and minimizes intra-subject variations. To guarantee its performance for better transfer learning, we utilize the popular Face-ResNet architecture [6] to build the convolutional neural network. We adopt the state-of-the-art Additive Maxmargin Softmax (AM-Softmax) loss function [23][4] [25] for training the base model.…”

Section: Training On Source Domainmentioning

confidence: 99%

“…On the other hand, face recognition technology has made tremendous strides in the past five years, mainly due to the availability of large scale face training data and deep neural network models for face recognition. Verification Rate (VR) on the Labeled Faces in the Wild (LFW) dataset, one of the first public domain "faces in the wild" dataset, at a False Accept Rate (FAR) of 0.1% has increased from 41.66% in 2014 [12] to 98.65% in 2017 [6]. Hence, the earlier published results on ID document photo to live face matching are now obsolete.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DocFace: Matching ID Document Photos to Selfies

Shi

Jain

2018

2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems (BTAS)

View full text Add to dashboard Cite

Numerous activities in our daily life, including transactions, access to services and transportation, require us to verify who we are by showing our ID documents containing face images, e.g. passports and driver licenses. An automatic system for matching ID document photos to live face images in real time with high accuracy would speed up the verification process and remove the burden on human operators. In this paper, by employing the transfer learning technique, we propose a new method, DocFace, to train a domain-specific network for ID document photo matching without a large dataset. Compared with the baseline of applying existing methods for general face recognition to this problem, our method achieves considerable improvement. A cross validation on an ID-Selfie dataset shows that DocFace improves the TAR from 61.14% to 92.77% at FAR=0.1%. Experimental results also indicate that given more training data, a viable system for automatic ID document photo matching can be developed and deployed.

show abstract

Section: Training On Source Domainmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

DocFace: Matching ID Document Photos to Selfies

Shi

Jain

2018

2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems (BTAS)

View full text Add to dashboard Cite

show abstract

“…Our discriminative models consist of a popular CNN architecture based on residual blocks that has been used in other face recognition works [61,62,63]. The network is trained with softmax loss and optimised using stochastic gradient descent with momentum.…”

Section: Methodsmentioning

confidence: 99%

Generating Photo-Realistic Training Data to Improve Face Recognition Accuracy

Trigueros¹,

Li²,

Hartnett³

2018

Preprint

View full text Add to dashboard Cite

In this paper we investigate the feasibility of using synthetic data to augment face datasets. In particular, we propose a novel generative adversarial network (GAN) that can disentangle identity-related attributes from non-identity-related attributes. This is done by training an embedding network that maps discrete identity labels to an identity latent space that follows a simple prior distribution, and training a GAN conditioned on samples from that distribution. Our proposed GAN allows us to augment face datasets by generating both synthetic images of subjects in the training set and synthetic images of new subjects not in the training set. By using recent advances in GAN training, we show that the synthetic images generated by our model are photo-realistic, and that training with augmented datasets can indeed increase the accuracy of face recognition models as compared with models trained with real images alone.

show abstract

“…Deep neural networks dominate the ongoing research in face recognition [34], [32], [28], [22], [18], [9], [24], [36], given their success in the ImageNet competition [14]. Taigman et al [34] proposed the first face recognition application using deep neural networks.…”

Section: Deep Face Recognitionmentioning

confidence: 99%

“…Training: We first train a base CNN model on a cleaned version 7 of MS-Celeb-1M dataset [7] to learn face recognition of still images, for which we adopts the Face-ResNet (DeepVisage) architecture [9]. The component-wise quality module is then trained on the same training dataset using the template triplet loss.…”

Section: Implementation Detailsmentioning

confidence: 99%

Video Face Recognition: Component-wise Feature Aggregation Network (C-FAN)

Gong

Shi

Kalka

et al. 2019

2019 International Conference on Biometrics (ICB)

View full text Add to dashboard Cite

We propose a new approach to video face recognition. Our component-wise feature aggregation network (C-FAN) accepts a set of face images of a subject as an input, and outputs a single feature vector as the face representation of the set for the recognition task. The whole network is trained in two steps: (i) train a base CNN for still image face recognition; (ii) add an aggregation module to the base network to learn the quality value for each feature component, which adaptively aggregates deep feature vectors into a single vector to represent the face in a video. C-FAN automatically learns to retain salient face features with high quality scores while suppressing features with low quality scores. The experimental results on three benchmark datasets, YouTube Faces [39], IJB-A [13], and IJB-S [12] show that the proposed C-FAN network is capable of generating a compact feature vector with 512 dimensions for a video sequence by efficiently aggregating feature vectors of all the video frames to achieve state of the art performance.

show abstract

DeepVisage: Making face recognition simple yet with powerful generalization skills

Cited by 13 publications

References 0 publications

DocFace: Matching ID Document Photos to Selfies

DocFace: Matching ID Document Photos to Selfies

Generating Photo-Realistic Training Data to Improve Face Recognition Accuracy

Video Face Recognition: Component-wise Feature Aggregation Network (C-FAN)

Contact Info

Product

Resources

About