Search

Over the past years, the main research innovations in face recognition focused on training deep neural networks on large-scale identity-labeled datasets using variations of multi-class classification losses. However, many of these datasets are retreated by their creators due to increased privacy and ethical concerns. Very recently, privacy-friendly synthetic data has been proposed as an alternative to privacy-sensitive authentic data to comply with privacy regulations and to ensure the continuity of face recognition research. In this paper, we propose an unsupervised face recognition model based on unlabeled synthetic data (USynthFace). Our proposed USynthFace learns to maximize the similarity between two augmented images of the same synthetic instance. We enable this by a large set of geometric and color transformations in addition to GAN-based augmentation that contributes to the USynthFace model training. We also conduct numerous empirical studies on different components of our USynthFace. With the proposed set of augmentation operations, we proved the effectiveness of our USynthFace in achieving relatively high recognition accuracies using unlabeled synthetic data., Comment: Accepted at Face and gesture 2023

In order to explore the failure characteristics and safety control measures of multiple mining soft rock roadway, and based on the fracture characteristics of mining soft rock roadway, drilling column and rock mechanics test results, FLAC3D is used to establish a numerical model to analyze the failure characteristics of the original support soft rock roadway under multiple mining, and the joint control measures of goaf filling and reinforcement support are proposed, field measurement and analysis of its treatment effect. The results show that: the failure of support is obvious, the vertical stress of mining face is transferred to the surrounding rock mass, and the stress concentration and deformation of the lower right side roadway are caused The filling of the goaf limits the stress concentration, displacement and plastic range of the surrounding rock to a certain extent, and the reinforcement support further strengthens the strength of the broken surrounding rock. The comprehensive treatment scheme of goaf filling and secondary reinforcement support has achieved good engineering results.

With the increased deployment of face recognition systems in our daily lives, face presentation attack detection (PAD) is attracting much 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 by variations in sensors/illuminations. The other stream leverages 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 demonstrate 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 problems. We successfully prove the design of our proposed PAD solution in a step-wise 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, Accepted at IEEE Winter Conference on Applications of Computer Vision (WACV 2022)

Face presentation attack detection (PAD) is critical to secure face recognition (FR) applications from presentation attacks. FR performance has been shown to be unfair to certain demographic and non-demographic groups. However, the fairness of face PAD is an understudied issue, mainly due to the lack of appropriately annotated data. To address this issue, this work first presents a Combined Attribute Annotated PAD Dataset (CAAD-PAD) by combining several well-known PAD datasets where we provide seven human-annotated attribute labels. This work then comprehensively analyses the fairness of a set of face PADs and its relation to the nature of training data and the Operational Decision Threshold Assignment (ODTA) on different data groups by studying four face PAD approaches on our CAAD-PAD. To simultaneously represent both the PAD fairness and the absolute PAD performance, we introduce a novel metric, namely the Accuracy Balanced Fairness (ABF). Extensive experiments on CAAD-PAD show that the training data and ODTA induce unfairness on gender, occlusion, and other attribute groups. Based on these analyses, we propose a data augmentation method, FairSWAP, which aims to disrupt the identity/semantic information and guide models to mine attack cues rather than attribute-related information. Detailed experimental results demonstrate that FairSWAP generally enhances both the PAD performance and the fairness of face PAD.

The problem of stoping on ultra-long working face is one of challenges for making coal mining safe and efficient. The strong mining effect of surrounding rock induced by long working face mining causes a number of coal or rock dynamic disasters to safe and efficient coal mining. For the purpose of in-depth study of the law of earth’s surface movement on ultra-long working face, this paper, with FLAC3D (numerical simulation software) used to build a numerical model of ultra-long working face, analyzes the law of roof-to-floor strata behaviors in coal bed during mining, and studies the weakening of roof-to-floor parameters in coal bed, bearing pressure and the law of roof caving and rock strata movement; Analyzes the face length effect resulted from the difference in face lengths. It has been recognized that the sphere of influence of bearing pressure changes from small to large and then to small, and changes in distribution of arch bottom and arch height at the plastic failure zone are further intensified over advancing of the working face, moreover, the movement of rock strata is subject to the cumulative influence of the advancing direction of the working face and the face length effect. The above-mentioned research results can serve as a theoretical basis for practical engineering.

The recent COVID-19 pandemic has increased the focus on hygienic and contactless identity verification methods. However, the pandemic led to the wide use of face masks, essential to keep the pandemic under control. The effect of wearing a mask on face recognition (FR) in a collaborative environment is a currently sensitive yet understudied issue. Recent reports have tackled this by evaluating the masked probe effect on the performance of automatic FR solutions. However, such solutions can fail in certain processes, leading to performing the verification task by a human expert. This work provides a joint evaluation and in-depth analyses of the face verification performance of human experts in comparison to state-of-the-art automatic FR solutions. This involves an extensive evaluation by human experts and 4 automatic recognition solutions. The study concludes with a set of take-home messages on different aspects of the correlation between the verification behavior of humans and machines., Comment: Accepted at the IET Biometrics journal (2022)

A novel hybrid fluorocarbon cationic surfmer has been synthesized and its aggregation and surface properties have been studied by surface tension, electrical conductivity, steady-state fluorescence, Rayleigh light scattering, dynamic light scattering and transmission electron microscopy. Through surface tension, electrical conductivity, steady-state fluorescence and Rayleigh light scattering measurements, the effectiveness of surface tension reduction, the maximum surface excess concentration, the minimum area occupied per surfactant molecule at the air/water interface, the micropolarity and aggregation number of micelles were investigated. The results shows that the surfmer has superior surface activity and lower micropolarity than other surfmers. The critical micelle concentration at different temperatures and a series of thermodynamic parameters (\(\Delta G_{\text{mic}}^{\text{o}}\), \(\Delta H_{\text{mic}}^{\text{o}}\), and \(\Delta S_{\text{mic}}^{\text{o}}\), \(\Delta G_{\text{ads}}^{\text{o}}\), \(\Delta H_{\text{mic}}^{\text{A}}\), and \(\Delta C_{\text{pmic}}^{\text{o}}\)) of micellization were evaluated. The thermodynamic parameters showed that the micelle formation was entropy-driven in the temperature range of 15–40 °C. The size and morphology of the aggregates were also confirmed by dynamic light scattering and transmission electron microscopy.