In Cross-Domain Few-Shot Learning (CD-FSL), models are required to identify novel classes while addressing domain discrepancies caused by visual style variations. Simple style transformations often fail to extend beyond the source domain’s distribution, and unrepresentative support samples in the target task may lead to ambiguous or biased decision boundaries. To address these challenges, a Style-Guided Source Data Augmentation and Target Feature Optimization (SSDATFO) approach is proposed. Specifically, Style-Guided Source Data Augmentation is introduced, employing Style Transformation and Source Data Augmentation techniques to create more challenging source data, thereby expanding the source domain’s style distribution. Target Feature Optimization is subsequently introduced, comprising two distinct modules. The Domain Attention Shift Transformation enhances low-magnitude feature channels, thereby reactivating target domain feature channels previously overlooked by the source domain-trained feature extractor. Additionally, the Task Category Differentiation Enhancement Transformation calibrates the features of support samples and eliminates the commonality component along both the task-specific and inter-class commonality directions for all features within the novel task, thereby acquiring more discriminative features. Extensive experiments on eight distinct target datasets demonstrate the efficacy of the proposed method, while comprehensive ablation studies and detailed visualization experiments elucidate its nuanced and compelling aspects.

The collection of large-scale datasets inevitably introduces noisy labels, leading to a substantial degradation in the performance of deep neural networks (DNNs). Although sample selection is a mainstream method in the field of learning with noisy labels, which aims to mitigate the impact of noisy labels during model training, the testing performance of these methods exhibits significant fluctuations across different noise rates and types. In this paper, we propose Cross-to-Merge Training (C2MT), a novel framework that is insensitive to the prior information in sample selection progress, enhancing model robustness. In practical implementation, using cross-divided training data, two different networks are cross-trained with the co-teaching strategy for several local rounds, subsequently merged into a unified model by performing federated averages on the parameters of two models periodically. Additionally, we introduce a new class balance strategy, named Median Balance Strategy (MBS), during the cross-dividing process, which evenly divides the training data into a labeled subset and an unlabeled subset based on the estimated loss distribution characteristics. Extensive experimental results on both synthetic and real-world datasets demonstrate the effectiveness of C2MT. The Code will be available at: https://github.com/LanXiaoPang613/C2MT.

Abstract Demirci–Selçuk meet-in-the-middle (DS-MITM) attack is an effective method for cryptanalysis. As far as we know, the published automatic results of DS-MITM attack are all for byte-oriented ciphers. In this article, we first propose the automatic analysis method of DS-MITM attack for bit-oriented ciphers based on constraint programming, which is integrated with key-bridging technique. Based on the automatic modeling method, we propose the first result of DS-MITM attack on SIMON, which is a family of lightweight block ciphers proposed by the National Security Agency (NSA) in 2013.

Due to the emergence of heterogeneous Internet of Medical Things (IoMT) (e.g., wearable health devices, smartwatch monitoring, and automated insulin delivery systems), large volumes of patient data are dispatched to central cloud servers for disease analysis and diagnosis. Although this direct mode brings a lot of convenience for both patients and medical professionals (MPs), the open communication channel between them also incurs several security and privacy issues, such as man-in-the-middle attacks, eavesdropping attacks, and tracking attacks. Based on the unsolved challenges in wireless medical sensor networks (WMSNs), several researchers have proposed various authentication and key agreement (AKA) protocols for this type of healthcare system recently. However, most of these protocols do not perceive physical-layer security and over-centralized server problem in WMSN. In this article, to address these two open problems, we propose a lightweight and reliable authentication protocol for WMSN, which is composed of cutting-edge blockchain technology and physically unclonable functions (PUFs). In addition, a fuzzy extractor scheme is introduced to deal with biometric information. Subsequently, two security evaluation methods are used to prove the high reliability of our proposed scheme. Finally, performance evaluation experiments illustrate that the proposed mutual authentication protocol requires the least computation and communication cost among the compared schemes.

In this paper, we propose memory-efficient Generative Adversarial Nets (GANs) in line with knowledge distillation. Most existing GANs have a shortcoming in terms of the number of model parameters and low processing speed. Here, to tackle the problem, we propose Adversarial Knowledge Distillation for Generative models (AKDG) for highly efficient GANs, in terms of unconditional generation. Using AKDG, model size and processing speed are substantively reduced. Through an adversarial training exercise with a distillation discriminator, a student generator successfully mimics a teacher generator in fewer model layers and fewer parameters and at a higher processing speed. Moreover, our AKDG is network architecture-agnostic. A Comparison of AKDG-applied models to vanilla models suggests that it achieves closer scores to a teacher generator and more efficient performance than a baseline method with respect to Inception Score (IS) and Frechet Inception Distance (FID). In CIFAR-10 experiments, improving IS/FID 1.17pt/55.19pt and in LSUN bedroom experiments, improving FID 71.1pt in comparison to the conventional distillation method for GANs. Our project page is https://maguro27.github.io/AKDG/.

Convolutional neural networks (CNNs) have achieved great success in visual recognition because of the availability of large-scale image datasets, such as the ImageNet. The transfer of convolutional features to challenging scene recognition remains an open problem. Multiple non-linear transforms endow the convolutional features with abundant information. On the other side, CNNs are adept at capturing the holistic appearances of scenes, whereas the lack of some critical local details may reduce the recognition accuracy. To address these problems, we propose a novel hierarchical coding algorithm to learn effective representations. To adapt the scale variations, many useful patches with various scales sampled from the whole image are considered to provide the sufficient details. Non-negative sparse decomposition model (NNSD) based on convolutional features is proposed to learn the sharable components for each scale and further produce global signatures. Based on the global signatures, inter-class linear coding (ICLC) is proposed to learn the discriminative components and ultimate image representations. Experimental results indicate that our approach significantly improves the recognition accuracy compared with general CNN models and achieves excellent performance on five standard benchmarks.

The authors propose a method for real-time collaborative animation for 3D models with finger play and hand shadow. Two or more users have the model animated more complexly and expressively than one user. For instance, a model of a crocodile is animated by two users; one manipulates its mouth, neck and tail, and the other manipulates its four legs. With the implemented system up to five users can manipulate one model collaboratively. From evaluations following points were found: 1) users could manipulate models without detailed instruction; 2) most participants felt the system operable and enjoyable; 3) the motions made with the system were not less cute, amusing and lively than those by animators with conventional methods.

In this paper the authors report a method for animating 3D models with finger play and hand shadow. For preparation, the motion of the models is associated with a motion of the hands. Appropriate association based on finger play and hand shadow provides intuitive operation. For example, the user makes a hermit crab's claw pinch by pinching the index and middle finger. On using the system the user doesn't need to wear sensors or gloves. The operation of the method is so easy that children can animate 3D models. In the evaluation, participants are mostly positive to the method.

This paper presents a multi-input and multi-output (MIMO) ground penetrating radar (GPR) system, which is going to be launched to the moon for imaging shallow regolith structures and estimating the dielectric properties. This system, as an important part of China' Chang-E 5 lunar exploration mission, employs twelve off-ground Vivaldi antennas as transmitters/receivers, and works in a stationary mode. A reverse-time migration algorithm is developed to process the MIMO GPR dataset for obtaining a high-resolution image of the subsurface objects. The results of a laboratory experiment on a volcanic ash pit demonstrate that the upper and lower interfaces of a marble slab of 3 cm thickness buried at a depth up to 2 m can be clearly imaged. A full waveform inversion algorithm based on Born iterative method is applied to invert the dielectric properties of the subsurface objects. The preliminary results of a numerical experiment demonstrate that the dielectric permittivity of a subsurface cubic object can be accurately obtained using the MIMO GPR dataset at only six discrete frequencies.

This paper introduces an extensible SOA-based platform that facilitates implementation of various e-collaboration systems. First, it reviews the roadmap of SOA evolution and examines the trend of service orientation in network architecture field, then it constructs a holistic service-oriented platform integrating into Web services, component-based and messaging-based technologies. After analyzing the requirements of collaboration manufacturing (CM), a typical CM model for virtual manufacturing enterprise alliance is presented. It also defines four levels of CM services according to the degree of process-centric. Finally, it illustrates how to dynamically compose a service-oriented application solution using BPEL4WS

We have developed a very simple yet highly reliable face recognition method called the VQ histogram method. A codevector referred (or matched) count histogram, which is obtained by vector quantization (VQ) processing of the facial image, is utilized as a very effective personal feature. By applying appropriate low pass filtering and VQ processing to a facial image, useful features for face recognition can be extracted. Experimental results show a recognition rate of 95.6% for 400 images of 40 persons (10 images per person), which contain variations in lighting, pose, and expression, from the publicly available ORL database. Equal error rate (ERR) of 2.6% is obtained for the verification experiment. By combining multiple low pass filtering procedures, the recognition rate is increased to 97% or higher.