Unified unsupervised and semi-supervised domain adaptation network for cross-scenario face anti-spoofing

Jia, Yunpei; Zhang, Jie; Shan, Shiguang; Chen, Xilin

doi:10.1016/j.patcog.2021.107888

Cited by 50 publications

(19 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Face PAD Methods: In recent years, there has been an increasing number of studies in the field of face PAD [19] , [20] , [21] . These studies can be broadly grouped into three categories: texture-based methods, deep-learning-based methods, and hybrid methods.…”

Section: Related Workmentioning

confidence: 99%

Real masks and spoof faces: On the masked face presentation attack detection

Fang

Damer

Kirchbuchner

et al. 2022

Pattern Recognition

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Real masks and spoof faces: On the masked face presentation attack detection

Fang

Damer

Kirchbuchner

et al. 2022

Pattern Recognition

View full text Add to dashboard Cite

“…Though the generalization capacity has been improved by those techniques, they are still restricted to labeled data in source domains. Recently, several works [29,52] start introducing the unlabeled data into FAS training, while semi-supervised benchmarks with comprehensive scenarios (e.g., cross domains and attack types) have not been built yet.…”

Section: Related Workmentioning

confidence: 99%

Consistency Regularization for Deep Face Anti-Spoofing

Wang¹,

Yu²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Face anti-spoofing (FAS) plays a crucial role in securing face recognition systems. Empirically, given an image, a model with more consistent output on different views of this image usually performs better, as shown in Fig. 1. Motivated by this exciting observation, we conjecture that encouraging feature consistency of different views may be a promising way to boost FAS models.In this paper, we explore this way thoroughly by enhancing both Embedding-level and Prediction-level Consistency Regularization (EPCR) in FAS. Specifically, at the embedding-level, we design a dense similarity loss to maximize the similarities between all positions of two intermediate feature maps in a self-supervised fashion; while at the prediction-level, we optimize the mean square error between the predictions of two views. Notably, our EPCR is free of annotations and can directly integrate into semi-supervised learning schemes. Considering different application scenarios, we further design five diverse semisupervised protocols to measure semi-supervised FAS techniques. We conduct extensive experiments to show that EPCR can significantly improve the performance of several supervised and semi-supervised tasks on benchmark datasets. The codes and protocols will be released at https://github.com/clks-wzz/EPCR.

show abstract

“…In face anti-spoofing methods, according to the availability of target labels, DA can be used in two configurations to find a shared feature space between seen and unseen data distribution: unsupervised [49,78,91,76,100,109,95,116,115], semi-supervised [75,109]. These two are of particular importance because of the availability of partially labeled target data (semi-supervised) or lack of full annotation in target domain (unsupervised) for real-world applications.…”

Section: Domain Adaptationmentioning

confidence: 99%

“…Alternatively, Jia et al [109] proposed a DA method that can be applied in both semi-supervised and unsupervised manner according to target domain. Their structure consists of two main components; Marginal Distribution Alignment (MDA) module and Conditional Distribution Alignment (CDA) module.…”

Section: Domain Adaptationmentioning

confidence: 99%

“…• Dependence on source domain: domain adaptation methods attempt to close the gap between source and target domain, but as we can see in inter-dataset evaluation in table 5, domain adaptation methods' performance significantly vary when source and target domain are interchanged. For instance, in the inter-dataset test between CASIA-MFSD and Replay Attack, the methods have a dramatic performance drop when it is trained on Replay Attack unless the model has access to a few labeled data from target domain (semi-supervised case in [109]), which may not be feasible in many scenarios.…”

Section: Acer = Ap Cer + Bp Cermentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning meets Liveness Detection: Recent Advancements and Challenges

Sabaghi¹,

Oghbaie²,

Kooshan³

et al. 2021

Preprint

View full text Add to dashboard Cite

Facial biometrics has been recently received tremendous attention as a convenient replacement for traditional authentication systems. Consequently, detecting malicious attempts has found great significance, leading to extensive studies in face anti-spoofing (FAS),i.e., face presentation attack detection. Deep feature learning and techniques, as opposed to hand-crafted features, have promised dramatic increase in the FAS systems' accuracy, tackling the key challenges of materializing realworld application of such systems. Hence, a new research area dealing with development of more generalized as well as accurate models is increasingly attracting the attention of the research community and industry. In this paper, we present a comprehensive survey on the literature related to deep-feature-based FAS methods since 2017. To shed light on this topic, a semantic taxonomy based on various features and learning methodologies is represented. Further, we cover predominant public datasets for FAS in a chronological order, their evolutional progress, and the evaluation criteria (both intra-dataset and inter-dataset). Finally, we discuss the open research challenges and future directions.

show abstract

Unified unsupervised and semi-supervised domain adaptation network for cross-scenario face anti-spoofing

Cited by 50 publications

References 14 publications

Real masks and spoof faces: On the masked face presentation attack detection

Real masks and spoof faces: On the masked face presentation attack detection

Consistency Regularization for Deep Face Anti-Spoofing

Deep Learning meets Liveness Detection: Recent Advancements and Challenges

Contact Info

Product

Resources

About