Page 78 - profile2012.indd

Research Faculty 研究人員

助研究員 研究員
陳郁方 Yu-Fang Chen 陳祝嵩 Chu-Song Chen

Assistant Research Fellow Research Fellow
Ph.D, Informational Management, National Taiwan University Ph.D., Computer Science and Information Engineering, National Taiwan University

Tel: +886-2-2788-3799 ext. 1514 Fax: +886-2-2782-4814 Tel: +886-2-2788-3799 ext. 1310 Fax: +886-2-2782-4814
Email: yfc@iis.sinica.edu.tw Email: song@iis.sinica.edu.tw
http://www.iis.sinica.edu.tw/pages/yfc/index_en.html http://www.iis.sinica.edu.tw/pages/song

● Assistant Research Fellow, IIS, Academia Sinica (2009 - ) ● EATCS (European Association for Theoretical Computer Science) ● Deputy Director, CITI, Academia Sinica (2008- ) ● Secretary-General, IPPR Society (2007-2008)
● Adjunct Assistant Professor, IM, National Taiwan University (2010 - ) award for best theoretical paper at ETAPS 2010. ● Research Fellow, IIS, Academia Sinica (2008- ) ● Associate Research Fellow, IIS, Academia Sinica (2003-2008)
● Ph.D., IM, National Taiwan University (2009) ● Adjunct Professor, INM, National Taiwan University (2009- ) ● Assistant Research Fellow, IIS, Academia Sinica (1999-2003)
● PostDoc, Uppsala University (2009) ● Editorial Board Member, Machine Vision and Applications, ● Ph.D, CSIE, National Taiwan University (1996)
Springer (2010- )
Research Description Publications Research Description Publications

My research interests include automatic techniques that 1. Parosh Abdulla, Mohamed Faouzi Atig, Yu-Fang Chen, Carl My main research directions include computer vision, 1. Hsin-Chien Huang, Yung-Yu Chuang and Chu-Song Chen,
can be used to ensure program correctness w.r.t. desired Leonardsson and Ahmed Rezine, “Counter-Example Guided pattern recognition and image/signal processing. I have “Multiple Kernel Fuzzy Clustering,” IEEE Transactions on
th
properties. In particular, I’m interested in model checking Fence Insertion under TSO”, to appear in the 18 International worked on some fundamental and application-oriented Fuzzy Systems, volume 20, number 1, pages 120-134, Febru-
ary 2012.
and related theories such as automata and logic. The veri- Conference on Tools and Algorithms for the Construction and problems and contributed to the frontier knowledge in
Analysis of Systems (TACAS 2012).
cation problem of programs is in general undecidable. these areas. 2. Kuang-Yu Chang, Chu-Song Chen, and Yi-Ping Hung, “Or-

Intuitively, the problem involves several di erent domains 2. Parosh Abdulla, Yu-Fang Chen, Lorenzo Clemente, Lukas In recent ve years, my research interests focused on ma- dinal Hyperplanes Ranker with Cost Sensitivities for Age Es-
timation,” IEEE Computer Society Conference on Computer

of in niteness (e.g., unbounded data structure, the num- Holik, Chih-Duo Hong, Richard Mayr, Tomas Vojnar, “Ad- chine learning and the associated techniques applicable Vision and Pattern Recognition, CVPR 2011, June 2011.

vanced Ramsey-based Buchi Automata Inclusion Testing,”
ber of threads involved, the domain of data variables) and The 22 International Conference on Concurrency Theory to visual surveillance, human-computer interaction, and 3. Yu-Ting Chen, Chu-Song Chen, “Fast Human Detection Us-
nd
therefore a systematic general procedure to explore all (CONCUR 2011). image/video analyses. The research items include image ing A Novel Boosted Cascading Structure with Meta Stages,”

possible execution traces or program states is di cult to 3. Y.-F. Chen, E. Clarke, A. Farzan, M.-H. Tsai, Y.-K. Tsay, enhancement, aesthetic quality assessment of images/ IEEE Transactions on Image Processing, volume 17, number
obtain. However, for certain classes of programs and cor- and B.-Y. Wang, “Automated Assume-Guarantee Reasoning videos, age and gender estimation, lighting normaliza- 8, pages 1452-1464, August 2008.

rectness properties of interest, e cient veri cation pro- through Implicit Learning,” 22 International Conference on tion, pedestrian detection, motion segmentation, salient 4. Wen-Yan Chang, Chu-Song Chen, and Yi-Ping Hung, “Ana-

nd
cedures has been found. Normally such procedure will in- Computer Aided Verification (CAV 2010), Lecture Notes in feature matching, and so on. lyzing Facial Expression by Fusing Manifolds,” Asian Con-

volve both deep theoretical results and also huge amount Computer Science, July 2010. ference on Computer Vision, ACCV 2007, Lecture Notes in
Computer Science, 4844, November 2007.

of engineering e orts. 4. P.A. Abdulla, Y.-F. Chen, L.o Clemente, L. Holik, C.-D. Hong, In the future, my research topics are still toward image/
R. Mayr, and T. Vojnar, “Simulation Subsumption in Ramsey- video understanding from machine learning techniques. 5. Wen-Yan Chang, Chu-Song Chen, and Yi-Ping Hung, “Ap-
In the past few years, We have developed several tech- based B uchi Automata Universality and Inclusion Testing,” We plan to tackle the problems of building codewords for pearance-Guided Particle Filtering for Articulated Hand
niques that can be used to improve the performance of 22 International Conference on Computer Aided Verifi cation sparse representation in only a single pass of data scan, Tracking,” IEEE Conference on Computer Vision and Pattern
nd
model checking. To name a few, we developed a machine (CAV 2010), Lecture Notes in Computer Science, July 2010. employing multiple a nities/kernels for spectral cluster- Recognition, CVPR 2005, San Diego, CA, USA, June 2005.

learning based approach that can automatically divide 5. P.A. Abdulla, Y.-F. Chen, L. Holik, R. Mayr and T. Vojnar, ing, and learning proper projections for structural hash- 6. Chu-Song Chen and Wen-Yan Chang, “On Pose Recovery for

the veri cation problem of a systems into several sub- “When Simulation Meets Antichain (on Language Inclusion ing. In addition, we will devote to the application of real- Generalized Visual Sensors,” IEEE Transactions on Pattern
problems on verifying components of the system. This Checking of NFA/TA) (Best Paper Award),” The 16 Interna- time photo aesthetics estimation for hand-held devices, Analysis and Machine Intelligence, volume 26, number 7,
th
pages 848-861, July 2004.
improved the performance of model checking on several tional Conference on Tools and Algorithms for the Construc- and also the development of ying/ oating cameras that

di cult problems. We also proposed a technique “simula- tion and Analysis of Systems (TACAS 2010), Lecture Notes in can follow people’s movement and interact with people 7. Chu-Song Chen, Yu-Ting Chen, and Fay Huang, “Stitching

and Reconstruction of Linear-Pushbroom Panoramic Images
tion subsumption” that can be used to speed up the deci- Computer Science, March 2010. by hand postures/gestures. for Planar Scenes,” European Conference on Computer Vi-
sion procedure of automata language inclusion, which is 6. Parosh Abdulla , Yu-Fang Chen, Giorgio Delzanno, Frederic sion, ECCV 2004, May 2004.

a frequently used procedure in program veri cation. This Haziza, Chih-Duo Hong and Ahmed Rezine., “Constrained 8. Chu-Song Chen, Yi-Ping Hung, and Jen-Bo Cheng,
technique helped us to earn a best ETAPS paper award Monotonic Abstraction: a CEGAR for Parameterized Veri- “RANSAC-based DARCES: A New Approach to Fast Auto-
st

fication,” The 21 International Conference on Concurrency
from EATCS (European Association for Theoretical Com- Theory (CONCUR 2010), Lecture Notes in Computer Sci- matic Registration of Partially-Overlapping Range Images,”
puter Science). We also have a procedure to ensure that a ence, 2010. IEEE Transactions on Pattern Analysis and Machine Intelli-
distributed algorithm is correct w.r.t some safety proper- 7. Y.-K. Tsay, Y.-F. Chen, M.-H. Tsai, K.-N. Wu, W.-C. Chan, gence, volume 21, number 11, pages 1229-1234, November
1999.
ties regardless of the threads involved. We are working on C.-J. Luo, and J.-S. Chang, “Tool Support for Learning Buechi
extending this approach to support programs under weak Automata and Linear Temporal Logic,” Formal Aspects of 9. Chu-Song Chen, Chi-Kuo Yu, and Yi-Ping Hung, “New Cal-
ibration-free Approach for Augmented Reality Based on Pa-
memory model. We now also develop an approach that Computing, volume 21, pages 259-275, 2009. rameterized Cubed Structure,” International Conference on

can support veri cation of programs with pointers, recur- 8. Y.-F. Chen, A. Farzan, E.M. Clarke, Y.-K. Tsay, and B.-Y. Computer Vision, ICCV 1999, pages 30-37, September 1999.
sion and multi-thread. Wang, “Learning Minimal Separating DFA’s for Composition- 10. Chu-Song Chen, Yi-Ping Hung, Chiann-Chu Chiang, and Ja-
al Verification,” The 15 International Conference on Tools Ling Wu, “Range Data Acquisition Using Color Structured

th
and Algorithms for the Construction and Analysis of Systems Lighting and Stereo Vision,” Image and Vision Computing,
(TACAS 2009), Lecture Notes in Computer Science, 5505, volume 15, pages 445-456, June 1997.
pages 31-45, 2009.
研究人員
78 Research Faculty
79