Huiyun Mao

Huiyun Mao

Auto-generated profile, based on publicly available information

Huiyun Mao

College of Computer Science, South China University of Technology, Guangzhou, China

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 September 2019 in Sensors
Verified
Journal Article
Open access
PDF file
LLLei Li
Lei Li
MGMian Guo
Mian Guo
LMLihong Ma
Lihong Ma
HMHuiyun Mao
Huiyun Mao
QGQuansheng Guan
Quansheng Guan
Fog computing has recently emerged as an extension of cloud computing in providing high-performance computing services for delay-sensitive Internet of Things (IoT) applications. By offloading tasks to a geographically proximal fog computing server instead of a remote cloud, the delay performance can be greatly improved. However, some IoT applications may still experience considerable delays, including queuing and computation delays, when huge amounts of tasks instantaneously feed into a resource-limited fog node. Accordingly, the cooperation among geographically close fog nodes and the cloud center is desired in fog computing with the ever-increasing computational demands from IoT applications. This paper investigates a workload allocation scheme in an IoT–fog–cloud cooperation system for reducing task service delay, aiming at satisfying as many as possible delay-sensitive IoT applications’ quality of service (QoS) requirements. To this end, we first formulate the workload allocation problem in an IoT-edge-cloud cooperation system, which suggests optimal workload allocation among local fog node, neighboring fog node, and the cloud center to minimize task service delay. Then, the stability of the IoT-fog-cloud queueing system is theoretically analyzed with Lyapunov drift plus penalty theory. Based on the analytical results, we propose a delay-aware online workload allocation and scheduling (DAOWA) algorithm to achieve the goal of reducing long-term average task serve delay. Theoretical analysis and simulations have been conducted to demonstrate the efficiency of the proposal in task serve delay reduction and IoT-fog-cloud queueing system stability.

23 December 2019 in IEEE Access
Verified
Journal Article
Open access
TATasweer Ahmad
Tasweer Ahmad
HMHuiyun Mao
Huiyun Mao
LLLuojun Lin
Luojun Lin
GTGuozhi Tang
Guozhi Tang
Human skeleton contains significant information about actions, therefore, it is quite intuitive to incorporate skeletons in human action recognition. Human skeleton resembles to a graph where body joints and bones mimic to graph nodes and edges. This resemblance of human skeleton to graph structure is the main motivation to apply graph convolutional neural network for human action recognition. Results show that the discriminant contribution of different joints is not equal for different actions. Therefore, we propose to use attention-joints that correspond to joints significantly contributing to the specific actions. Features corresponding to only these attention-joints are computed and assigned as node features of the graph. In our method, node features (also termed as attention-joint features) include the i) distances of attention-joints from the center-of-gravity of human body, ii) distances between adjacent attention-joints and iii) joints flow features. The proposed method gives a simple but more efficient representation of skeleton sequences by concatenating more relative distances and relative coordinates to other joints. The proposed methodology has been evaluated on single image Stanford 40-Actions dataset, as well as on temporal skeleton-based action recognition PKU-MDD and NTU-RGBD datasets. Results show that this framework outperforms existing state-of-the-art methods.

23 December 2019 in IEEE Access
Verified
Journal Article
Open access
SZSheng Zhang
Sheng Zhang
GTGuozhi Tang
Guozhi Tang
YLYuliang Liu
Yuliang Liu
HMHuiyun Mao
Huiyun Mao
Recently, deep learning has greatly promoted the performance of license plate recognition (LPR) by learning robust features from numerous labeled data. However, the large variation of wild license plates across complicated environments and perspectives is still a huge challenge to the robust LPR. To solve the problem, we propose an effective and efficient shared adversarial training network (SATN) in this paper, which can learn the environment-independent and perspective-free semantic features from wild license plates with the prior knowledge of standard stencil-rendered license plates, as standard stencil-rendered license plates are independent of complicated environments and various perspectives. Besides, to correct the features of heavily perspective distorted license plates perfectly, we further propose a novel dual attention transformation (DAT) module in the shared adversarial training network. Comprehensive experiments on AOLP-RP and CCPD benchmarks show that the proposed method outperforms state-of-the-art methods by a large margin on the LPR task.

4 August 2015
Verified
Book Chapter
Closed access
PDF file
HMHuiyun Mao
Huiyun Mao
WPWeishen Pan
Weishen Pan