Search

Face anti-spoofing (FAS), as an important means to protect face recognition models, can ensure that the system remains secure and reliable in the face of various presentation attacks. The current deep learning-based face anti-spoofing model has satisfactory results when the test data and training data obey the same distribution, but the accuracy of the model decreases considerably when the trained model infers in the scene outside the domain, such as cross-domain transfer and out-of-distribution scenarios. The problems that silent face anti-spoofing models will encounter in real scenarios, i.e., the models encounter unknown environments and unknown attack methods, are mainly described. The corresponding solutions are classified into four categories: methods based on domain adaptation, methods based on domain generalization, methods based on zero shot or few shot learning, and methods based on anomaly detection. Each solution and its deep learning model methods are summarized and compared. The mechanism, network structure, advantages, limitations and application scenarios of some major methods are summarized. After that, common public datasets, evaluation metrics, measurement protocols commonly used for face anti-spoofing in out of domain scenarios and test results of state-of-the-art methods under some protocols are introduced. Finally, the difficulties and challenges of face anti-spoofing in practical applications are discussed, and future research directions are summarized.

Have environmental regulations introduced by local governments in China promoted regional green development? This paper identifies the impact and spillover effect of environmental regulations on Green Total Factor Productivity by using provincial panel data over the period 2005-2018. We found that the intensity of ERs is positively correlated with regional GTFP growth and also benefits its neighbours. I have also identified green innovation and FDI transfer as two mechanisms of action leading to this phenomenon. Finally, the effects of environmental regulation also vary according to geographical location, the degree of dependence on polluting industries and the level of nationalisation.

To clarify the influence mechanism of initial defects on the internal crack evolution law and damage destruction of cement superfine tailings backfill(CSTB), uniaxial compression and simultaneous acoustic emission (AE) tests were carried out on CSTB with different degrees of internal initial defects constructed by doping with different contents of doping agent(DA), and the AE b-value, amplitude fractal dimensional time series characteristics and three-dimensional distribution law of AE events of CSTBs were analyzed in this paper. The research shows that the doping of DA caused the fluctuation and jump of the CSTB b-value to be significantly enhanced, and the b-value showed two jumps in the loading process. The fractal characteristics of CSTB with different DA doping are better, and the fractal dimension D-value and b-value both show a rapid decreasing trend in the preparation process of the specimen for the occurrence of overall unstable failure. The admixture of DA increases the number of initial defects and non-homogeneity within the CSTB, making the generation of 3D localization points of AE events somewhat random. The results of the study can help to better analyze the unstable failure process of the backfill in the mine hollow area and carry out the corresponding support work. [ABSTRACT FROM AUTHOR]

Studying the mechanical characteristics of hydrate-bearing sediments (HBS) contributes to the comprehensive understanding of the mechanical behavior in environments with natural gas hydrate (NGH) occurrences. Simultaneously, the distribution patterns of hydrates significantly influence the strength, deformation, and stability of HBS. Therefore, this paper employs particle flow code (PFC) to conduct biaxial discrete element simulations on specimens of HBS with different hydrate distribution patterns, revealing the macroscale–mesoscale mechanical properties, evolution patterns, and destructive mechanisms. The results indicate that the strain-softening behavior of HBS specimens strengthens with the increase in hydrate layer thickness, leading to higher peak strength and E50 values. During the gradual movement of the hydrate layer position (Ay) from both ends to the center of the specimen (Ay = 0.40 mm → Ay = 20 mm), the strain-softening behavior weakens. However, when Ay = 20 mm, the specimen exhibits evident strain-softening behavior again. Moreover, with an increase in the angle between the hydrate layer and the horizontal direction (α) greater than 20°, the peak strength of the specimen increases, while E50 shows an overall decreasing trend. The influence of axial loads on the hydrate layer in specimens varies with α, with larger contact forces and fewer cracks observed for higher α values. [ABSTRACT FROM AUTHOR]