A System for Disguised Face Recognition with Convolution Neural Networks

Law

et al. 2022

Vision

We investigated whether Father Christmas has a distinguishable facial phenotype by performing a cross-sectional cohort study examining the facial feature vectors of all publicly available photographs obtained from a google image search of individuals meeting our eligibility criteria presenting as Father Christmas compared with other adult and elderly bearded men. Facial feature vectors were determined using the open-source OpenFace facial recognition system and assessed by support vector machines (SVM). SVM classifiers were trained to distinguish between the facial feature vectors from our groups. Accuracy, precision, and recall results were calculated and the area under the curve (AUC) of the receiver operating characteristic (ROC) were reported for each classifier. SVM classifiers were able to distinguish the face of Father Christmas from other adult men with a high degree of accuracy and could discriminate Father Christmas from elderly bearded men but with lower accuracy. Father Christmas appears to have a distinct facial phenotype when compared to adult men and elderly bearded men. This will be reassuring to children who may be keen to recognise him but raises some interesting questions about the careful use of two-dimensional facial analysis, particularly when employed to explore the relationships between genotype and facial phenotype in a clinical dysmorphology setting.

Section: Consideration Of Biasmentioning

confidence: 85%

Does Father Christmas Have a Distinctive Facial Phenotype?

Law

et al. 2022

Vision

“…The high classification accuracy of the DNN was somewhat surprising, because the system was not trained to recognise masked faces, which is typically a challenging task for face recognition systems (Hung et al 2018). Because the DNN is a black-box system, we can only infer how it processes masked faces by looking at the similarity ratings for each condition.…”

Section: Dnn Performancementioning

confidence: 99%

“…This face recognition system is a deep neural network (DNN) that was developed by the University of Surrey, which we had access to through the FACER2VM project. 3 Importantly, the DNN was not trained to identify masked or occluded faces, which has previously proven to be a challenging task for naïve face recognition systems (Dhamecha et al 2014;Hung et al 2018). Our aim in testing the DNN is to see whether any impairment to human performance would be mirrored in the performance of the naïve computer system.…”

Section: Introductionmentioning

confidence: 99%

Surgical face masks impair human face matching performance for familiar and unfamiliar faces

Carragher¹,

Hancock²

2020

Preprint

In response to the COVID-19 pandemic, many governments around the world now recommend, or require, their citizens to cover the lower half of their face in public. Consequently, many people now wear surgical face masks in public. We investigated whether surgical face masks affected the performance of human observers, and a state-of-the-art face recognition system, on tasks of perceptual face matching. Participants judged whether two simultaneously presented face photographs showed the same person or two different people. We superimposed images of surgical masks over the faces, creating three different mask conditions: control (no masks), mixed (one face wearing a mask), and masked (both faces wearing masks). We find that surgical face masks have a large detrimental effect on human face matching performance, and that the degree of impairment is the same regardless of whether one or both faces in each pair are masked. Surprisingly, this impairment is similar in size for both familiar and unfamiliar faces. When matching masked faces, human observers are biased to reject unfamiliar faces as mismatches and to accept familiar faces as matches. Finally, the face recognition system showed very high classification accuracy for control and masked stimuli, even though it had not been trained to recognise masked faces. However, accuracy fell markedly when one face was masked and the other was not. Our findings demonstrate that surgical face masks impair the ability of humans, and naïve face recognition systems, to perform perceptual face matching tasks. Matching decisions for masked faces should be treated with caution.

CONTECSI International Conference on Information Systems and Technology Management

“…No trabalho de (Hung, 2018) é relatado um sistema de reconhecimento facial, para identificar indivíduos, com algum tipo de disfarce (óculos, chapéu, cachecol, etc. ), baseado em Rede Neural Convolucional Normalizada Profunda (DNCNN).…”

Section: Trabalhos Relacionadosunclassified

Analysis and Evaluation of Digital Image Enhancement Techniques Applied to Facial Recognition With Neural Networks.

SOUZA¹,

ARAÚJO²

Machine Learning algorithms are used for automating detection such as Artificial Neural Networks. São utilizados softwares que filtram a imagem usando técnicas de Processamento de Imagem afim de definir padrões e simplificar a busca, agilizando dessa forma, o processamento. Transforming the resulting image into a binary vector. This vector is used as input in a Neural Network to extract face patterns and perform facial detection. The proposal is promising with hits above 80% with the Gaussian filter and reaching 90% with the Histogram Equalization filter that allow the proposed methodology is used regardless of the filter applied to the image. This work proposes methodology for facial recognition. After performing the image treatment and using filters such as Gaussian and Histogram Equalization, algorithms are applied for transforming the resulting image into a binary vector. They are increasingly used in biometric authentication applications where facial image is used to identify people previously registered. In some cases, facial image can also locate suspects in a crowd or can even assist in the search for missing persons.