Camera Invariant Feature Learning for Generalized Face Anti-Spoofing

Chen, Baoliang; Yang, Wenhan; Li, Haoliang; Wang, Shiqi

doi:10.1109/tifs.2021.3055018

Cited by 37 publications

(8 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on this evidence, Wang et al [101] propose to supervise the FAS patch models via an asymmetric angular-margin softmax loss to relax the intra-class constraints among PAs. On the other hand, to provide more confident predictions on hard samples, Chen et al [137] adopt the binary focal loss to guide the model to enlarge the margin between live/spoof samples and achieve strong discrimination for hard samples.…”

Section: Direct Supervision With Binary Cross Entropy Lossmentioning

confidence: 99%

Deep Learning for Face Anti-Spoofing: A Survey

Qin

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

Face anti-spoofing (FAS) has lately attracted increasing attention due to its vital role in securing face recognition systems from presentation attacks (PAs). As more and more realistic PAs with novel types spring up, early-stage FAS methods based on handcrafted features become unreliable due to their limited representation capacity. With the emergence of large-scale academic datasets in the recent decade, deep learning based FAS achieves remarkable performance and dominates this area. However, existing reviews in this field mainly focus on the handcrafted features, which are outdated and uninspiring for the progress of FAS community. In this paper, to stimulate future research, we present the first comprehensive review of recent advances in deep learning based FAS. It covers several novel and insightful components: 1) besides supervision with binary label (e.g., '0' for bonafide vs. '1' for PAs), we also investigate recent methods with pixel-wise supervision (e.g., pseudo depth map); 2) in addition to traditional intra-dataset evaluation, we collect and analyze the latest methods specially designed for domain generalization and open-set FAS; and 3) besides commercial RGB camera, we summarize the deep learning applications under multi-modal (e.g., depth and infrared) or specialized (e.g., light field and flash) sensors. We conclude this survey by emphasizing current open issues and highlighting potential prospects.

show abstract

Section: Direct Supervision With Binary Cross Entropy Lossmentioning

confidence: 99%

Deep Learning for Face Anti-Spoofing: A Survey

Qin

et al. 2022

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

show abstract

“…An element-wise weighing fusion strategy was followed in this model. In [ 26 ], the authors used camera-invariant feature learning while focusing on generalisation in FPAD. This framework learned both high-frequency and low-frequency information.…”

Section: Related Workmentioning

confidence: 99%

Fusion Methods for Face Presentation Attack Detection

Abdullakutty

Johnston

Elyan

2022

Sensors

View full text Add to dashboard Cite

Face presentation attacks (PA) are a serious threat to face recognition (FR) applications. These attacks are easy to execute and difficult to detect. An attack can be carried out simply by presenting a video, photo, or mask to the camera. The literature shows that both modern, pre-trained, deep learning-based methods, and traditional hand-crafted, feature-engineered methods have been effective in detecting PAs. However, the question remains as to whether features learned in existing, deep neural networks sufficiently encompass traditional, low-level features in order to achieve optimal performance on PA detection tasks. In this paper, we present a simple feature-fusion method that integrates features extracted by using pre-trained, deep learning models with more traditional colour and texture features. Extensive experiments clearly show the benefit of enriching the feature space to improve detection rates by using three common public datasets, namely CASIA, Replay Attack, and SiW. This work opens future research to improve face presentation attack detection by exploring new characterizing features and fusion strategies.

show abstract

“…Frequency domain analysis is a classical method in image signal processing and has been widely used for general image classification or texture classification task [25,11]. Moreover, some face PAD methods attempted [16,4,5] to transform the images in frequency domain and mine the artifacts cues. The results showed that features in frequency domain is less sensitive to the variations of the capture environments (e.g., sensors or light conditions).…”

Section: Multi-level Frequency Decomposition (Mfd)mentioning

confidence: 99%

“…In addition to widely used LBP features, several studies [16,4,5] attempted to transform images to the frequency domain. Li et al [16] utilized the dissimilarity in Fourier spectra by considering that less high frequency components exist in attacks compared to bona fide samples.…”

Section: Introductionmentioning

confidence: 99%

Learnable Multi-level Frequency Decomposition and Hierarchical Attention Mechanism for Generalized Face Presentation Attack Detection

Fang¹,

Damer²,

Kirchbuchner³

et al. 2021

Preprint

View full text Add to dashboard Cite

With the increased deployment of face recognition systems in our daily lives, face presentation attack detection (PAD) is attracting a lot of attention and playing a key role in securing face recognition systems. Despite the great performance achieved by the hand-crafted and deep learning based methods in intra-dataset evaluations, the performance drops when dealing with unseen scenarios. In this work, we propose a dual-stream convolution neural networks (CNNs) framework. One stream adapts four learnable frequency filters to learn features in the frequency domain, which are less influenced variations in sensors/illuminations. The other stream leverage the RGB images to complement the features of the frequency domain. Moreover, we propose a hierarchical attention module integration to join the information from the two streams at different stages by considering the nature of deep features in different layers of the CNN. The proposed method is evaluated in the intra-dataset and cross-dataset setups and the results demonstrates that our proposed approach enhances the generalizability in most experimental setups in comparison to state-of-the-art, including the methods designed explicitly for domain adaption/shift problem. We successfully prove the design of our proposed PAD solution in a stepwise ablation study that involves our proposed learnable frequency decomposition, our hierarchical attention module design, and the used loss function. Training codes and pre-trained models are publicly released 1 .

show abstract

Camera Invariant Feature Learning for Generalized Face Anti-Spoofing

Cited by 37 publications

References 53 publications

Deep Learning for Face Anti-Spoofing: A Survey

Deep Learning for Face Anti-Spoofing: A Survey

Fusion Methods for Face Presentation Attack Detection

Learnable Multi-level Frequency Decomposition and Hierarchical Attention Mechanism for Generalized Face Presentation Attack Detection

Contact Info

Product

Resources

About