B.Sc.(CUHK) M.Phil.(CUHK) Ph.D.(CUHK)
Chi-Sing Leung has
been working
on image-based rendering and neural networks for 20 years. He has published over 60
international journal papers and 10 book chapters. He received the IEEE Transactions on Multimedia 2005 Prize Paper Award for the
paper titled, The Plenoptic Illumination Function in IEEE Trans. on Multimedia, Vol. 4, Issue
3, September, 2002. He proposed a two-level
compression mjthod for illumination adjustable images and the
corresponding real-time rendering methods. He has written several game
development related articles in Shader
X3, Shader X4,
and Shader X6 . Since 2000,
he has been providing programming consultancy services to Development Bank of
Singapore (Hong Kong). In 2007, he gave a
lecture, Is there anything
comparable to spherical harmonics but simpler?, in Game Developers Conference 2007 San
Francisco. He is also a governing board member of the Asian Pacific Neural
Network Assembly (APNNA).
Phone: (852)-3442 7378 Email: eeleungc@cityu.edu.hk
Research Interests
I would like to recruit graduate students who have strong
interests in pursuing their Ph.D. in the area of machine learning, neural
networks, signal processing, computer graphics, and computer vision. Potential
candidates should possess a first-class honor (or average marks >85 or GPA
> 3.5) master
degree in Electronic Engineering, Computer Engineering or other relevant
disciplines.
If you are interested, please send me an email with a brief resume.
I would like to recruit a postdoctoral fellow
who has strong background on mathematics and statistics. It is expected that
the postdoctoral fellow will perform research on fault tolerant neural networks.
Potential candidates should possess a Ph.D. in Electronic Engineering, Computer
Engineering or other relevant disciplines.
If you are interested, please send me an email with a brief resume. Interested
candidates from mainland can apply for the Hong
Kong Scholars Program.
1
|
Chi Sing Leung and John Sum, “RBF Networks Under the Concurrent Fault
Situation,” accepted
for publication in IEEE Transactions on
Neural Networks and Learning Systems [Previous name is IEEE Transactions
on Neural Networks]. |
2
|
John Sum, Chi Sing Leung and K. Ho, “Convergence analyses on on-line
weight noise injection-based training algorithms for MLPs,” accepted for publication in IEEE Transactions on Neural Networks and
Learning Systems [Previous name is IEEE Transactions on Neural Networks]. |
3
|
Yi Xiao , Yuxin Liu, Chi-Sing Leung, John Sum, K. Ho, “Analysis on the
Convergence Time of Dual Neural Network-Based KWTA,” IEEE Transactions on Neural Networks and Learning Systems, 23
(4), 676-682, April 2012. [Previous name is IEEE Transactions on Neural
Networks] |
4
|
Yi Xiao,
Liang Wan, Chi-Sing Leung, Yu-Kun Lai, and Tien-Tsin
Wong, “Example-based Color Transfer for Gradient Meshes,” accepted for
publication in IEEE Transactions on
Multimedia. |
5
|
John Sum, Chi-Sing Leung, K. Ho, “On-Line Node Fault Injection
Training Algorithm for MLP Networks: Objective Function and Convergence
Analysis ,” IEEE Transactions on Neural
Networks and Learning Systems, 23 (2), 211-222, Feb 2012. [Previous name
is IEEE Transactions on Neural
Networks] |
6
|
Liang Wan, Shue-Kwan
Mak, Tien-Tsin Wong,
Chi-Sing Leung, “Spatio-Temporal Sampling of
Dynamic Environment Sequences,” IEEE
Transactions on Visualization and Computer Graphic, 17 (10), pp.
1499-1509, Oct 2011. |
7
|
K. Ho, Chi Sing Leung, and J. Sum, J., “Objective
functions of online weight noise injection training algorithms for MLPs,” IEEE Transactions on Neural Networks,
22(2), pp 317-323, Feb 2011. |
8
|
T.Y. Ho, L.
Wan, Chi Sing Leung, P.M. Lam, and T.-T. Wong, “Unicube
for dynamic environment mapping,” IEEE Transactions on Visualization and
Computer Graphics, 17(1), pp 51-63, Jan 2011. |
9
|
K. HO, Chi Sing Leung, and John SUM, “Convergence and Objective Functions of Some Fault/Noise-Injection-Based Online Learning,” IEEE Transactions on Neural Networks, vol. 21, 6, JUNE 2010, 938-947. |
10
|
Chi Sing Leung, H.J. Wang, and J. Sum, “On the selection of weight decay parameter for faulty networks,” IEEE Transactions on Neural Networks, 21(8), pp 1232-1244, Aug 2010. |