QLQingbao Li

Qingbao Li

*This is an algorithmically generated author record based on publicly available information.

Qingbao Li has a total of 22 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A CAN Bus Anomaly Detection Method Based on Time Series Prediction

    by Weiping Yao, Qingbao Li, Zhifeng Chen, Bocheng Xu

  • UEFI Security Threats Introduced by S3 and Mitigation Measure

    by Weihua Jiao, Qingbao Li, Zhifeng Chen, Fei Cao

    UEFI is widely used as the new generation BIOS, except for some new features, it also brings some security issues. The threat brought by S3 sleep jeopardizes the security of computer platforms seriously. An attacker can disable the write protection of UEFI by tampering with the S3 boot script or execute arbitrary code by exploiting dispatch opcodes with the entry point out of SMRAM. Therefore, it is vital to solve the security issue related to S3 and propose effective mitigation measures. Based on the S3 related modules and the risk propagation between modules, we propose the MBPE algorithm to identify the key vulnerability of the platform through formal analysis. Aiming at this critical vulnerability, based on the LockBox protection mechanism, we propose a mitigation measure. It includes changing the storage time of the boot script, the changed storage address of the boot script and a specific UEFI variable. After taking this mitigation measure, the attacker cannot obtain the boot script and its location simultaneously. Such, it can protect the boot script from being tampered with and further attackers. Finally, based on the real attack cases, we evaluate the mitigation measure by the abstraction of attack path and mathematical methods. Based on the evaluation finding, the mitigation measures can prevent S3related attacks and effectively improve platform security.

  • Open Access

    3D Face Image Inpainting with Generative Adversarial Nets

    by Tongxin Wei, Qingbao Li, Jinjin Liu, Ping Zhang, Zhifeng Chen

    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.


  • Open Access

    CP-GAN: A Cross-Pose Profile Face Frontalization Boosting Pose-Invariant Face Recognition

    by Jinjin Liu, Qingbao Li, Ming Liu, Tongxin Wei

    Pose variant or self-occlusion is one of the open issues which severely degrades the performance of pose-invariant face recognition (PIFR). Existing solutions to PIFR either have undesirable generalization based on challenging pose normalization or are complicated for implement on account of deep neural network. To relieve the impact of ill-pose on PIFR, we have proposed Cross-Pose Generative Adversarial Networks(CP-GAN) to frontalize the profile face with unaltered identity by learning the mapping between the profile and frontal faces in image space. The generator is an encoder-decoder U-net, and generate frontal face image by fusing multiple profile images to achieve a better performance in PIFR. The siamese discriminative network attends to extract the deep representations of the generated frontal face and the ground truth without introducing extra networks in verification and recognition. Besides the implementable architecture, this problem is well alleviated by introducing a combination of adversarial loss for both the generator and the discriminator, symmetry loss, patch-wise loss, and identity loss guiding an identity reserving property of the generated frontal view. Quantitative and qualitative evaluation on both controlled and in-the-wild datasets attest that the solution we proposed to PIFR presents satisfactory perceptual results and transcends state-of-the-art methods on ill-pose face recognition.


  • Open Access

    Reverse Engineering of Intel Microcode Update Structure

    by Zhiguo Yang, Qingbao Li, Ping Zhang, Zhifeng Chen

    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.


  • Open Access

    PA-GAN: A Patch-Attention Based Aggregation Network for Face Recognition in Surveillance

    by Ming Liu, Jinjin Liu, Ping Zhang, Qingbao Li

    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.


  • Open Access

    Research on Security Detection Technology for Internet of Things Terminal Based on Firmware Code Genes

    by Xinbing Zhu, Qingbao Li, Zhifeng Chen, Guimin Zhang, Peng Shan

    Internet of Things (IoT) terminals have firmware with heterogeneous, closed-source, and heavy business but light security characteristics, whereas on the edge, there are limited resources and a high code reuse rate. Once there are security risks at the firmware level, these risks are difficult to detect and discover, and the resulting impact quickly spreads over a wide range. Therefore, a similarity and homology analysis of firmware codes in an IoT terminal will be helpful for further research on firmware malicious code detection, vulnerability mining, backdoor discovery and copyright protection. Inspired by biological genes, this paper attempts to break away from the traditional feature-centered approach and focuses on code classification and the qualitative description of code features to discuss the idea of code similarity and homology analysis. Additionally, the proposed approach is information-centric, focusing on the informativeness (essentiality, stability, antivariability, and heritability) of the firmware code genes and the quantitative analysis of firmware code similarity and homology by discussing common methods and mechanisms. This paper presents security detection technology for IoT terminal firmware by measuring the gene distance between the codes. A prototype firmware security detection system (FSDS) for IoT terminals based on firmware code genes is designed and implemented. The experimental results show that this method has a good search matching effect and has certain advantages over traditional firmware security detection methods based on similarity theory.


  • Open Access

    Efficient Cache Strategy for Face Recognition System

    by Guoxiao Zong, Qingbao Li, Ping Zhang, Guimin Zhang, Xinbing Zhu

    Recently, the need for real-time deep learning applications, such as face recognition system, implement on the embedded devices is increasing. At the system level, caching system is one of the most effective ways to reduce response time. However, embedded front-end devices can hardly guarantee effective system overhead for complex cache system while deploying deep learning model. In this paper, this paper presented an efficient cache strategy for deep learning applications such as face recognition system. This paper first established a distributed embedded system model to describe to functions of the face recognition systems. Then, this paper figures out the goal and idea of optimizing the cache design theoretically. Finally, our cache strategy is experimented and verified better than the traditional methods.


  • Open Access

    Unpaired Domain Transfer for Data Augment in Face Recognition

    by Jinjin Liu, Qingbao Li, Ping Zhang, Guimin Zhang, Ming Liu

    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.


  • Open Access

    A Firmware Code Gene Extraction Technology for IoT Terminal

    by Xinbing Zhu, Qingbao Li, Ping Zhang, Zhifeng Chen

    With the development of the IoT technology, an unprecedented number of IoT terminals are connected to various networks. Commercial-off-the-shelf(COTS) technology is widely used in the IoT terminal firmware, which results in high code reuse rates. Such firmware is always heterogeneous and closed-source. It is so difficult to detect and investigate the security risks at the firmware level that their impacts are faster and broader. In recent years, some firmware security detection technologies based on similarity are gradually becoming a research hotspot. However, in these studies, the basic issue regarding whether these foundations comprise an essential basis for comparison and their utility as similarity measures has not been addressed theoretically. Inspired by biological genes, this paper attempts to supplement a foundation for cross-platform firmware binary code homology and similarity analysis by mining firmware code genes that can essentially identify code and exhibit stability, antivariability and heritability. The firmware code gene extract system(FCGES) is designed and implemented in this paper. FCGES first extracts the features of firmware code, then numericizes and normalizes them, and finally sublimates them to firmware code genes by the hypothesis margin. The experimental results show that the firmware code gene extracted by FCGES has essentiality, stability, antivariability and heritability on different platforms.