24 February 2020 in IEEE Access
Face recognition is one of the hottest issues in the field of computer vision and pattern recognition. Deep learning-based recognition models already have more exceptional recognition ability than the human being on open datasets, but still cannot fully undertake the identity recognition task in real scenarios without human assistance. In this paper, we mainly analyze two obstacles, i.e., domain gap and training data shortage. We propose the unpaired Domain Transfer Generative Adversarial Network (DT-GAN) to relieve these two obstacles. We improve the GAN baseline to bridge the domain gap among datasets by generating images conforming to the style of a target domain by learning the mapping between the source domain and target domain. The generator could synthesize face with an arbitrary viewpoint at the same time. The model is trained with a combination of style transfer loss, identity loss, and pose loss, which ensures the successive domain transfer and data augment. We conduct experiments to testify the effectiveness and reasonability of DT-GAN. Experimental results demonstrate the recognition performance is dramatically boosted after domain transfer and data augment.
19 August 2020 in IEEE Access
Face recognition in unconstraint surveillance is a complicated problem on account of motion blur, expression variations and low resolution. Recent works have demonstrated that patch-attention is strictly more powerful than convolution in recognition models. In this study, we investigate the task of unconstraint surveillance face recognition. First, a Patch-Attention Generative Adversarial Network (PA-GAN) model is devised to aggregate some robust features on behalf of a set of raw surveillance frames, which not only increases the recognition accuracy but also reduces the computational costs of face matching. Second, an improved center loss function combined with abundant unlabeled surveillance faces is utilized to accurately classify the known identities. With the proposed method, the discriminativeness of the face representations is largely enhanced. Finally, the proposed method is verified in two widely used datasets, IJB-A dataset and QMUL-SurvFace dataset to demonstrate the effectiveness. Evaluation of the algorithm performances in comparison with other state-of-the-art methods indicates that the proposed design can achieve competitive accuracy on both the verification and identification protocols.
14 December 2020 in Mathematical Problems in Engineering
In the process of face recognition, face acquisition data is seriously distorted. Many face images collected are blurred or even missing. Faced with so many problems, the traditional image inpainting was based on structure, while the current popular image inpainting method is based on deep convolutional neural network and generative adversarial nets. In this paper, we propose a 3D face image inpainting method based on generative adversarial nets. We identify two parallels of the vector to locate the planer positions. Compared with the previous, the edge information of the missing image is detected, and the edge fuzzy inpainting can achieve better visual match effect. We make the face recognition performance dramatically boost.
1 February 2020 in Computers & Security
1 March 2019 in Computers & Security
15 September 2020 in IEEE Access
Microcode update mechanism have been widely used in modern processors. Due to the implementation details are not public, researchers are prevented from gaining any sort of further understanding currently. The microcode update binary which uploaded into Central Processing Unit (CPU) is the only accessible node in this update chain by researchers, but previous manual reverse analysis for a small amount of microcode updates has the disadvantages of incomplete coverage, slow speed, and low accuracy. Therefore, we first build a Sample Repository containing 504 Intel official microcode updates, then propose a semiautomatic analytical method named SJNW-MA to analyze samples. This work has the following merits: (1) automatic methods of similarity analysis and candidate feature mining improve the speed; (2) manual-assisted analysis based on expert knowledge can filter important features, to avoid redundant features or valuable common data blocks missing; (3) analysis for 504 microcode updates make the results of reverse engineering are more complete. Finally, we extract eleven structures of Intel microcode updates and group them into four categories. In addition, we also identify and describe some new metadata in microcode updates of the third and the fourth category, including a new 3072-bit RSA Modulus as well as corresponding RSA Exponent which indicates upgrade of security technology inside update mechanism.