Abstract Presentation attacks are becoming a serious threat to one of the most common biometric applications, namely face recognition (FR). In recent years, numerous methods have been presented to detect and identify these attacks using publicly available datasets. However, such datasets are often collected in controlled environments and are focused on one specific type of attack. We hypothesise that a model’s accurate performance on one or more public datasets does not necessarily guarantee generalisation across other, unseen face presentation attacks. To verify our hypothesis, in this paper, we present an experimental framework where the generalisation ability of pre-trained deep models is assessed using four popular and commonly used public datasets. Extensive experiments were carried out using various combinations of these datasets. Results show that, in some circumstances, a slight improvement in model performance can be achieved by combining different datasets for training purposes. However, even with a combination of public datasets, models still could not be trained to generalise to unseen attacks. Moreover, models could not necessarily generalise to a learned format of attack over different datasets. The work and results presented in this paper suggest that more diverse datasets are needed to drive this research as well as the need for devising new methods capable of extracting spoof-specific features which are independent of specific datasets. 

https://link.springer.com/content/pdf/10.1007/s12559-022-10037-z.pdf

Deep Transfer Learning on the Aggregated Dataset for Face Presentation Attack Detection

The invention relates to a living body detection method and device, computer equipment and a storage medium. The method comprises the steps that training set images are input into a living body detection model, a target living body detection model is obtained through training, and the living body detection model comprises a texture network, an optical flow network and a brightness network; the to-be-detected image is input into the target living body detection model, the living body detection result is output, and the face image is analyzed in parallel through the three feature networks, so that on one hand, a user does not need to perform instruction interaction, the detection efficiency of living body detection is improved, and on the other hand, the robustness of living body detection is improved.

Living body detection method and device, computer equipment and storage medium

AbstractThe robustness and generalization ability of Presentation Attack Detection (PAD) methods is critical to ensure the security of Face Recognition Systems (FRSs). However, in a real scenario, Presentation Attacks (PAs) are various and it is hard to predict the Presentation Attack Instrument (PAI) species that will be used by the attacker. Existing PAD methods are highly dependent on the limited training set and cannot generalize well to unknown PAI species. Unlike this specific PAD task, other face related tasks trained by huge amount of real faces (e.g. face recognition and attribute editing) can be effectively adopted into different application scenarios. Inspired by this, we propose to trade position of PAD and face related work in a face system and apply the free acquired prior knowledge from face related tasks to solve face PAD, so as to improve the generalization ability in detecting PAs. The proposed method, first introduces task specific features from other face related task, then, we design a Cross-Modal Adapter using a Graph Attention Network (GAT) to re-map such features to adapt to PAD task. Finally, face PAD is achieved by using the hierarchical features from a CNN-based PA detector and the re-mapped features. The experimental results show that the proposed method can achieve significant improvements in the complicated and hybrid datasets, when compared with the state-of-the-art methods. In particular, when training on the datasets OULU-NPU, CASIA-FASD, and Idiap Replay-Attack, we obtain HTER (Half Total Error Rate) of 5.48% for the testing dataset MSU-MFSD, outperforming the baseline by 7.39%. The source code is available at https://github.com/WentianZhang-ML/FRT-PAD.KeywordsFacePresentation attack detectionGraph neural network 

/pdf/effective-presentation-attack-detection-driven-by-face-vl1ie1cu.pdf

Effective Presentation Attack Detection Driven by Face Related Task

Adaptive lighting systems can be designed to detect the spatial characteristics of the visual environment and adjust the light output to increase visual comfort and performance. Such systems would require computational metrics to estimate occupants’ visual perception of indoor environments. This paper describes an experimental study to investigate the relationship between the perceived quality of indoor environments, personality, and computational image quality metrics. Forty participants evaluated the visual preference, clarity, complexity, and colorfulness of 50 images of indoor environments. Twelve image quality metrics (maximum local variation (MLV), spatial frequency slope (α), BRISQUE, entropy (S), ITU spatial information (SI), visual complexity (Rspt), colorfulness (M), root mean square (RMS) contrast, Euler, energy (E), contour, and fractal dimension) were used to estimate participants’ subjective evaluations. While visual clarity, visual complexity, and colorfulness could be estimated using at least one metric, none of the metrics could estimate visual preference. The results indicate that perceived colorfulness is highly correlated with perceived clarity and complexity. Personality traits tested by the 10-item personality inventory (TIPI) did not impact the subjective evaluations of the indoor environmental images. Future studies will explore the impact of target and background luminance on the perceived quality of indoor images.

/pdf/image-quality-metrics-personality-traits-and-subjective-1tv9m93b.pdf

Image Quality Metrics, Personality Traits, and Subjective Evaluation of Indoor Environment Images

Exploring Masked Image Modeling for Face Anti-spoofing

With the rapid development in face recognition, most of the existing systems can perform very well in unconstrained scenarios. However, it is still a very challenging task to detect face spoofing attacks, thus face anti-spoofing has become one of the most important research topics in the community. Though various anti-spoofing models have been proposed, the generalisation capability of these models usually degrades for unseen attacks in the presence of challenging appearance variations, e.g. , background, illumination, diverse spoofing materials and low image quality. To address this issue, we propose to use a Generative Adversarial Network (GAN) that transfers an input face image from the RGB domain to the depth domain. The generated depth clue enables biometric preservation against challenging appearance variations and diverse image qualities. To be more specific, the proposed method has two main stages. The first one is a GAN-based domain transfer module that converts an input image to its corresponding depth map. By design, a live face image should be transferred to a depth map whereas a spoofing face image should be transferred to a plain (black) image. The aim is to improve the discriminative capability of the proposed system. The second stage is a classification model that determines whether an input face image is live or spoofing. Benefit from the use of the GAN-based domain transfer module, the latent variables can effectively represent the depth information, complementarily enhancing the discrimination of the original RGB features. The experimental results obtained on several benchmarking datasets demonstrate the effectiveness of the proposed method, with superior performance over the state-of-the-art methods. The source code of the proposed method is publicly available at https://github.com/coderwangson/DFA .

From RGB to Depth: Domain Transfer Network for Face Anti-Spoofing

The invention discloses a living body detection method based on a novel pooling and attention mechanism double-flow network. The method comprises the following steps: constructing a double-flow network combining a spatial flow network and a time flow network; extracting an optical flow picture and carrying out face detection to obtain two inputs of the double-flow network; adding spatial pyramid pooling and global average hybrid pooling to the tail end of the double-flow network, classifying the two pooled features by using a full connection layer, and performing score-level fusion; carrying out feature-level fusion on the spatial flow network and the time flow network; and finally outputting the fused features to complete living body detection. The method has the beneficial effects that the time dimension is fully considered, and the proposed spatial pyramid and global average hybrid pooling strategy can effectively utilize features; for a data set containing various attack types andhaving large picture quality differences, the network model can obtain a low error rate.

Wang Yahang

Papers

From RGB to Depth: Domain Transfer Network for Face Anti-Spoofing

Living body detection method based on novel pooling and attention mechanism double-flow network