Group role assignment (GRA) maximizes total benefits by assigning agents to appropriate roles, while GRA with a training plan (GRATP) further considers the impact of training. However, existing research on GRA and GRATP does not fully consider the demand for flexible adjustment of human resource assignment, which may lead to increased employment costs and project delays. Moreover, role importance significantly affects training resource assignment, as key roles contribute more to overall performance. Therefore, we propose the GRA with secondment and role-importance-based training plan (GRA-SRIT) model to address these issues. Specifically, this article introduces seconded personnel to temporarily replace the positions of agents undergoing training, ensuring the smooth continuation of the project. Depending on the role requirements, different training durations are assigned based on the specific requirements of their roles. In addition, trainers with different levels of expertise are assigned to agents based on role importance, ensuring that critical roles receive more specialized training, thus maximizing total benefit. Finally, experiments demonstrate the proposed model’s effectiveness in different scenarios.

Data uncertainty refers to the degree of uncertainty in model predictions caused by data variability, challenging the model robustness in multimedia applications. The increased variability of uncontrolled face images, such as mask occlusion and image blur, increases the intra-class differences and inter-class similarities. As a result, the ambiguity in the learned features exacerbates the uncertainty of sample-to-class membership. Traditional face recognition models are deterministic point embedding models that fail to measure data uncertainty. Probabilistic embedding models, such as advanced Data Uncertainty Learning (DUL), represent each face image as a Gaussian distribution to measure data uncertainty. However, these models perform random sampling from the distribution once per training, and the sampled points may fall into different class regions, leading to training oscillation. Therefore, we propose a robust Region Uncertainty Learning (RUL) method, which adopts the entire Gaussian distribution of each sample during each training epoch, and estimates the region relations between the sample distribution region and the class region to measure the sample-to-class membership. In fact, DUL is a special case of the proposed RUL. Specifically, DUL estimates point-with-region relations and only represents absolute membership and non-membership. In contrast, RUL estimates region-with-region relations, enabling it to additionally represent incomplete membership. This more comprehensive membership measurement fully represents the uncertainty of membership, enhancing the model performance and robustness in uncontrolled scenes. Furthermore, for robust face recognition, we propose two RUL-based angular margin losses, AngleFace and RegionFace, to adaptively adjust the learning weights according to the uncertainty of membership. Finally, we comprehensively evaluate the effectiveness of RUL on various face datasets, and profoundly analyze the role of region relations. In future, we will explore the applicability of RUL in other tasks.

Since mask occlusion causes plentiful loss of facial feature, Masked Face Recognition (MFR) is a challenging image processing task, and the recognition results are susceptible to noise. However, existing MFR methods are mostly deterministic point embedding models, which are limited in representing noise images. Moreover, Data Uncertainty Learning (DUL) fails to achieve reasonable performance in MFR. Therefore, we propose a novel two-stream convolutional network, masked face data uncertainty learning (MaskDUL), that solves the problems by sampling uncertainty and intra-class distribution learning in MFR. Specifically, a Hard Kullback-Leibler Divergence (H-KLD) method is proposed to serve as an adaptive variance regularizer and a magnitude-based module is adopted to adaptively adjust the angular margin of different samples. Finally, insightful evaluation demonstrates the effectiveness and robustness of our MaskDUL.

In the post-epidemic era, Masked Face Recognition (MFR) is of great significance to our daily life, but it confronts a severe challenge of lacking real-world large-scale masked face datasets with identity labels.Moreover, mask enhances the diversity of face images and further improves the requirements for datasets.To address the above problem, we propose a novel CycleGAN-based masked face generation method MaskedFace-GAN (MFGAN), which is able to generate correct, authenticlooking and type-diverse masked face while ensuring the invariance of facial features.We design a three-stage training pipeline for MFGAN, which corresponds to three modules, respectively.Specifically, a facial feature detector is adopted to guide the model to generate the correct mask in the correct position.Then, by utilizing a mask binary segmentation module, the authenticity of generated images can be guaranteed.Lastly, with mask style encoder, the model can be optimized towards generating typediverse masked faces.Finally, comparing with advanced masked face synthesis and generation methods comprehensively, our MFGAN achieves the best results.Then we apply the generated masked face datasets to MFR model training, which further proves the feasibility of training MFR models on generated datasets and the effectiveness and advancement of MFGAN compared with other state-of-the-art methods.

High‐efficiency satellite‐ground downlink data transmission system is being demanded to support increasing volume of data with limited spectrum resource. Considering the dynamic link conditions due to the orbital motion of low Earth orbit (LEO) satellite, variable coding modulated orthogonal frequency division multiplexing (VCM‐OFDM) is proposed as an effective way to better exploit link resource. In this study, regarding the relative motion between the satellite on‐board transmitter and the ground station receiver, filtered‐OFDM signal is designed with adjustable coding modulation sets to achieve higher spectral efficiency and get along with the high‐Doppler environment at the same time. Physical layer frame structure is proposed taking the Doppler frequency shift compensation into account with minimum overhead. System performance evaluations prove that transmission throughput in the dynamic communication link can be enhanced significantly by the proposed scheme, which indicates the potential for supporting future gigabit X‐band reliable satellite‐ground downlink transmissions.

Linear least squares (LLS) estimation is a suboptimum but low-complexity localization method based on measurements of location-related parameters. It has been proved that selection of the reference anchor influences the LLS localization accuracy. In addition, hybridization of different types of measurements can fix the deficiencies of one type of measurements. In this paper, we proposed a new reference selection criterion for the hybrid TOA/RSS LLS localization technique (called H-LLSRS), which considers both measured ranges and the information about their coarse variances. Moreover, we consider a general scenario that variances of range measurements are different, and derive a weighted LLS (WLLS) estimator for hybrid TOA/RSS localization according to the information about the accurate ranging variances and the correlations among the observations. Simulation results show that if the RSS-based ranging variances are considerably larger than the TOA-based ranging variances, the H-LLS-RS localization technique yields better accuracy than the conventional LLS localization techniques. Furthermore, reference selection has no effect on the accuracy of WLLS localization technique.