Search

To address the problem of reduced face recognition accuracy in masked scenarios, this paper proposes a masked face reconstruction algorithm DeMaskGAN, which uses the Transformer Reconstruction Head (TRH) to restore the masked face features, and uses the Transformer Segmentation Head as an aid so that the TRH focuses on the masked face region and reconstructs the face to an unmasked state while maintaining the identity information. To improve the model performance, identity consistency, key point consistency, and perceptual consistency supervision mechanisms for faces are proposed to assist in training the model, and data augmentation methods are used to generate Mask-FFHQ datasets adapted to the mask-obscured face segmentation and reconstruction tasks, the experimental results show that the reconstructed face images enable the face recognition algorithm MobileFaceNet to achieve an AUC metric of 0.9743, which is 0.039 better than the direct use of MobileFaceNet to recognize masked faces. [ABSTRACT FROM AUTHOR]

Most research in the field of radar signal processing focuses on the use of time-frequency images (TFIs) to distinguish between different signal types. However, most studies have only examined the TFIs of a single signal, making it challenging to analyze and process the simultaneous reception of multiple signal components. This study proposes the use of adversarial latent separation auto encoder to separate and recognize multi-component signals, and innovatively propose a multi-network structure of feature extraction sub-network and signal separation sub-network. Thus, the problem of multi-component signal recognition is solved. Following separation, each component retains its time-frequency data while removing the influence of other components, and the separated TFIs are then subjected to parameter estimation and structural similarity (SSIM) measurements. The experimental findings demonstrate that the parameters retrieved from the separated signal have a low error with respect to the original signal, especially at low signal-to-noise ratios. The excellent SSIM and parameter estimation metrics between the separation results and the time-frequency image of the target tag imply that the separated single-component signal can be successfully reconstructed. [ABSTRACT FROM AUTHOR]

Intelligent defect detection technology combined with deep learning has gained widespread attention in recent years. However, the small number, and diverse and random nature, of defects on industrial surfaces pose a significant challenge to deep learning-based methods. Generating defect images can effectively solve this problem. This paper investigates and summarises traditional defect generation and deep learning-based methods. It analyses the various advantages and disadvantages of these methods and establishes a benchmark through classical adversarial networks and diffusion models. The performance of these methods in generating defect images is analysed through various indices. This paper discusses the existing methods, highlights the shortcomings and challenges in the field of defect image generation, and proposes future research directions. Finally, the paper concludes with a summary. [ABSTRACT FROM AUTHOR]

Over the past few years, a variety of membership inference attacks against deep learning models have emerged, raising significant privacy concerns. These attacks can easily infer whether a sample exists in the training set of the target model with little adversary knowledge, and the inference accuracy is often much higher than random guessing, which causes serious privacy leakage. To this end, defenses against membership inference attacks have attracted great interest. However, the current available defense methods such as regularization, differential privacy, and knowledge distillation are unable to balance the trade‐off between privacy and utility well. In this paper, we combine knowledge distillation and generative adversarial networks to propose a novel training framework that can effectively defend against membership inference attacks, called KD‐GAN. Extensive experiments show that our method implements an attack success rate of nearly 0.5 (random guesses) which can successfully defend against membership inference attacks without causing significant damage to model utility, and consistently outperforming other defense methods in the balance of privacy and utility. [ABSTRACT FROM AUTHOR]

With the rapid development of network technology, the Internet has accelerated the generation of network traffic, which has made network security a top priority. In recent years, due to the limitations of deep packet inspection technology and port number-based network traffic identification technology, machine learning-based network traffic identification technology has gradually become the most concerned method in the field of traffic identification with its advantages. As the learning ability of deep learning in machine learning becomes more substantial and more able to adapt to highly complex tasks, deep learning has become more widely used in natural language processing, image identification, and computer vision. Therefore, more and more researchers are applying deep learning to network traffic identification and classification. To address the imbalance of current network traffic, we propose a traffic identification model based on generating adversarial deep convolutional networks (GADCN), which effectively fits and expands traffic images, maintains a balance between classes of the dataset, and enhances the dataset stability. We use the USTC-TFC2016 dataset as training and test samples, and experimental results show that the method based on GADCN has better performance than general deep learning models. [ABSTRACT FROM AUTHOR]

In order to address the problems of traditional inpainting algorithm models, such as the inability to automatically identify the specific location of the area to be restored, the cost of inpainting and the difficulty of inpainting, and the problems of structural and texture discontinuity and poor model stability in deep learning-based image inpainting, this paper proposes an image inpainting based on a contextual coherent attention. This paper designs a network model based on generative adversarial networks. First, to improve the global semantic continuity and local semantic continuity of images in image inpainting, a contextual coherent attention layer is added to the network; second, to solve the problems of slow convergence and insufficient training stability of the model, a cross-entropy loss function is used; finally, the trained generator is used to repair images. The experimental results are compared using PSNR and SSIM metrics, compared with the traditional GAN model, our model has a 3.782dB improvement in peak signal-to-noise ratio and a 0.025% improvement in structural similarity. The experimental results show that the image inpainting method in this paper has better performance in terms of image edge processing, pixel continuity and overall image structure. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer-Aided Design & Computer Graphics / Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao is the property of Gai Kan Bian Wei Hui and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Mechanical intelligent fault diagnosis is an important method to accurately identify the health status of mechanical equipment and ensure its safe operation. With the advent of the "big data" era, it has become an inevitable tendency to choose different deep network models to improve the ability of data processing and classify faults. Meanwhile, in order to improve the generalization performances of fault diagnosis methods in different diagnosis scenarios, some fault diagnosis algorithms based on deep transfer learning have been developed. This paper introduces the concepts of deep transfer learning and explains the investigation motive. The advent in intelligent fault diagnosis of instances-based deep transfer learning, network-based deep transfer learning, mapping based deep transfer learning and adversarial-based deep transfer learning in recent years are summarized. Finally, we discuss the existing problems and development trend of deep transfer learning for intelligent fault diagnosis. This research has a positive significance for utilising deep transfer learning method in mechanical fault diagnosis. [ABSTRACT FROM AUTHOR]

Icon design is an important part of UI design, and a design task that designers often encounter. During the design process, it is important to highlight the function of icons themselves and avoid excessive similarity with similar icons, i.e., to have a certain degree of innovation and uniqueness. With the rapid development of deep learning technology, generative adversarial networks (GANs) can be used to assist designers in designing and updating icons. In this paper, we construct an icon dataset consisting of 8 icon categories, and introduce state-of-the-art research and training techniques including attention mechanism and spectral normalization based on the original StyleGAN. The results show that our model can effectively generate high-quality icons. In addition, based on the user study, we demonstrate that our generated icons can be useful to designers as design aids. Finally, we discuss the potential impacts and consider the prospects for future related research. [ABSTRACT FROM AUTHOR]

• A data enhancement method based on generative adversarial network is proposed. • The method combines the advantages of data augmentation and feature selection. • Case studies on a simulated case and two real industrial processes are provided. • Comparisons to state-of-the-arts show effectiveness of the proposed method. Soft sensor plays an important role in improving product quality; however, practical applications may often face with the problem of small sample size, which is challenging for developing data-driven models in terms of feature selection and good generalization. This paper proposes a data enhancement approach for small sample size data-driven problems based on generative adversarial networks integrated with maximum relevance minimum redundancy (MRMR). First, sample expansion is performed on the initial data by using a generative adversarial network. Second, irrelevant variables are eliminated by the MRMR and optimal features are obtained. Finally, neural networks-based soft sensor modeling is performed using the augmented dataset and the selected features. The proposed method is tested on a simulated penicillin case, an actual penicillin production case and an actual erythromycin production case. Experimental results show that the proposed method outperforms state-of-the-art existing methods, which verify the effectiveness and superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Icon design is an important part of UI design, and a design task that designers often encounter. During the design process, it is important to highlight the function of icons themselves and avoid excessive similarity with similar icons, i.e., to have a certain degree of innovation and uniqueness. With the rapid development of deep learning technology, generative adversarial networks (GANs) can be used to assist designers in designing and updating icons. In this paper, we construct an icon dataset consisting of 8 icon categories, and introduce state-of-the-art research and training techniques including attention mechanism and spectral normalization based on the original StyleGAN. The results show that our model can effectively generate high-quality icons. In addition, based on the user study, we demonstrate that our generated icons can be useful to designers as design aids. Finally, we discuss the potential impacts and consider the prospects for future related research.

Copyright of Journal of Central South University is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)