Chun-Fei Hsu

Chun-Fei Hsu

Auto-generated profile, based on publicly available information

Chun-Fei Hsu

Department of Electrical Engineering, Chung Hua University, Hsinchu 300, Taiwan, Republic of China

Research Focus & Metrics

Academic Career Length

20Years (2007-Present)

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

Representative Publications

View All Publications
24 June 2016 in International Journal of Fuzzy Systems
Verified
Journal Article
Closed access
PDF file
Chun-Fei HsuChun-Fei Hsu
Chun-Fei Hsu
TLTsu-Tian Lee
Tsu-Tian Lee

30 November 2009 in Expert Systems with Applications
Verified
Journal Article
Closed access
CHChun-Fei Hsu
Chun-Fei Hsu
CCChao-Ming Chung
Chao-Ming Chung
Chih-Min LinChih-Min Lin
Chih-Min Lin
CHChia-Yu Hsu
Chia-Yu Hsu


16 October 2008 in Fuzzy Sets and Systems
Verified
Journal Article
Closed access
CHChun-Fei Hsu
Chun-Fei Hsu
PLPing-Zong Lin
Ping-Zong Lin
TLTsu-Tian Lee
Tsu-Tian Lee
CWChi-Hsu Wang
Chi-Hsu Wang

4 May 2021 in IEEE Access
Verified
Journal Article
Open access
BWBing-Fei Wu
Bing-Fei Wu
Bo-Rui ChenBo-Rui Chen
Bo-Rui Chen
Chun-Fei HsuChun-Fei Hsu
Chun-Fei Hsu
Many facial landmark methods based on convolutional neural networks (CNN) have been proposed to achieve favorable detection results. However, the instability landmarks that occur in video frames due to CNNs are extremely sensitive to input image noise. To solve this problem of landmark shaking, this study proposes a simple and effective facial landmark detection method comprising a lightweight U-Net model and a dynamic optical flow (DOF). The DOF uses the fast optical flow to obtain the optical flow vector of the landmark and uses dynamic routing to improve landmark stabilization. A lightweight U-Net model is designed to predict facial landmarks with a smaller model size and less computational complexity. The predicted facial landmarks are further fed to the DOF approach to deal with the unstable shaking. Finally, a comparison of several common methods and the proposed detection method is made on several benchmark datasets. Experimental evaluations and analyses show that not only can the lightweight U-Net model achieve favorable landmark prediction but also the DOF stabilizing method can improve the robustness of landmark prediction in both static images and video frames. It should be emphasized that the proposed detection system exhibits better performance than others without requiring heavy computational loadings.

14 February 2007 in Neurocomputing
Verified
Journal Article
Closed access
CHChun-Fei Hsu
Chun-Fei Hsu
GCGuan-Ming Chen
Guan-Ming Chen
TLTsu-Tian Lee
Tsu-Tian Lee