Page 149 - My FlipBook
P. 149
tinguished Research Fellow 廖弘源 研
究
Mark Liao 人
員
Ph.D., Electrical Engineering, Northwestern University, United States Faculty
T +86-2-2788-3799 ext. 2203 or 1550 E liao@iis.sinica.edu.tw
F +886-2-2789-3309 W iwnwdewx._iies.nsi.hnticmal.edu.tw/pages/liao/
・ Honorary Chair Professor, National Chiao Tung University (2016-present)
・ IEEE Fellow (2013)
・ Academia Sinica Young Investigator Award (1998)
・ Academia Sinica Investigator Award (2010)
・ Distinguished Research Award, National Science Council (2003-2006, 2010-2016)
・ TECO Award 東元獎 (2016)
・ Associate Editor, IEEE Trans. on Multimedia (1998-2001), Image Processing
(2009-2013), Information Forensics and Security (2009-2012), ACM Computing
Surveys (2018-present)
Research Description
Beginning in 2018, my research team received a four-year AI-related project from the Ministry of Science and Technology (MOST)
(2018/1~2021/12). The MOST grants about $280,000 to this project each year. The original expertise of my team is multimedia signal and
information processing, and MOST encourages research teams with di erent training background to invest in arti cial intelligence related
theory or applied research from di erent aspects. In fact, the MOST had recognized that AI will be a very important direction in the future, if
no resources were invested, Taiwan's competitiveness in science or industry would inevitably decline. Due to the hasty preparation time, we
started with multimedia information processing combined with deep learning as the topic. The MOST is concerned that R & D teams will only
publish papers as their ultimate goal, and will seriously consider cooperation with Taiwan's industry as one of the evaluation criteria. One of
the most important criteria is "problems given by the industry, and solved by the academia." My research team then chose to cooperate with
the listed IC company ELAN. We hope to integrate respective software and hardware advantage, and use this to develop "smart city tra c
ow solutions" that are not only applicable to Taiwan, but also export internationally. We expect to complete the following systems in four
years: (1) A sub-system that can directly execute edge computing at intersections to compute various parameters of tra c ow; (2) A sub-
system that can use computer vision technology to compute other tra c ow parameters between adjacent intersections; and (3) A sub-
system that can make use of the parameters obtained above to trigger a reinforcement learning process, and then dynamically adjust all
tra c signs in a certain range.
Publications
1. C. Y. Wang, H. Y. Mark Liao, P. Y. Chen, and J. W. Hsieh, 6. Nick C. Tang, Yen-Yu Lin, Ming-Fang Weng, and Hong-Yuan Brochure 2020
"Enriching Variety of Layer-wise Learning Information by Mark Liao, "Cross-Camera Knowledge Transfer for Multiview
Gradient Combination," IEEE International Conference on People Counting," IEEE Transactions on Image Processing ,
Computer Vision Workshop (ICCVW) "Low Power Computer volume 24, number 1, pages 80-93, January 2015.
Vision.'', October 2019.
7. Y. L. Chen, C. T. Hsu, and H. Y. Mark Liao, "Simultaneous
2. Jen-Chun Lin, Wen-Li Wei, Tyng-Luh Liu, C.-C. Jay Kuo, and H. Tensor Decomposition and Completion Using Factor Priors,"
Y. Mark Liao, "Tell Me Where is Still Blurry: Adversarial Blurred IEEE Transactions on Pattern Analysis and Machine Intelligence,
Region Mining and Refining," 27th ACM Multimedia Conference volume 36, number 3, pages 577-591, March 2014.
(long paper), Nice, France, October 2019.
8. M. F. Weng, Y. Y. Lin, Nick C Tang, and H. Y. Mark Liao,
3. Jen-Chun Lin, Wen-Li Wei, Tyng-Luh Liu, Yi-Hsuan Yang, "Visual Knowledge Transfer among Multiple Cameras for People
Hsin-Min Wang, Hsiao-Rong Tyan, and Hong-Yuan Mark Counting with Occlusion Handling," 20th ACM Multimedia
Liao, "Coherent Deep-Net Fusion To Classify Shots In Concert Conference (long paper), October 2012.
Videos," IEEE Transactions on Multimedia, volume 20, number
11, pages 3123-3136, November 2018. 9. C. H. Ling, Y. M. Liang, C. W. Lin, Y. S. Chen, and H. Y. Mark
Liao, "Human Object Inpainting Using Manifold Learningbased
4. Jen-Chun Lin, Wen-Li Wei, James Yang, Hsin-Min Wang, and Posture Sequence Estimation," IEEE Trans. on Image Processing,
Hong-Yuan Mark Liao, "Automatic Music Video Generation volume 20, number 11, pages 3124-3135, November 2011.
Based on Simultaneous Soundtrack Recommendation and Video
Editing," 25th ACM Multimedia Conference (long paper), San 10. Hsueh-Yi Sean Lin, Hong-Yuan Mark Liao, and Ja-Chen
Jose, USA, October 2017. Lin, "Visual Salience-Guided Mesh Decomposition," IEEE
Transactions on Multimedia, volume 9, number 1, pages P46-P57,
5. P. C. Wen, W. C. Cheng, Y. S. Wang, H. K. Khu, Nick C. January 2007.
Tang, and H. Y. Mark Liao, "Court Reconstruction for Camera
Calibration in Broadcast Basketball Videos," IEEE Transactions
on Visualization and Computer Graphics, volume 22, number 5,
pages 1517-1526, May 2016.
147
究
Mark Liao 人
員
Ph.D., Electrical Engineering, Northwestern University, United States Faculty
T +86-2-2788-3799 ext. 2203 or 1550 E liao@iis.sinica.edu.tw
F +886-2-2789-3309 W iwnwdewx._iies.nsi.hnticmal.edu.tw/pages/liao/
・ Honorary Chair Professor, National Chiao Tung University (2016-present)
・ IEEE Fellow (2013)
・ Academia Sinica Young Investigator Award (1998)
・ Academia Sinica Investigator Award (2010)
・ Distinguished Research Award, National Science Council (2003-2006, 2010-2016)
・ TECO Award 東元獎 (2016)
・ Associate Editor, IEEE Trans. on Multimedia (1998-2001), Image Processing
(2009-2013), Information Forensics and Security (2009-2012), ACM Computing
Surveys (2018-present)
Research Description
Beginning in 2018, my research team received a four-year AI-related project from the Ministry of Science and Technology (MOST)
(2018/1~2021/12). The MOST grants about $280,000 to this project each year. The original expertise of my team is multimedia signal and
information processing, and MOST encourages research teams with di erent training background to invest in arti cial intelligence related
theory or applied research from di erent aspects. In fact, the MOST had recognized that AI will be a very important direction in the future, if
no resources were invested, Taiwan's competitiveness in science or industry would inevitably decline. Due to the hasty preparation time, we
started with multimedia information processing combined with deep learning as the topic. The MOST is concerned that R & D teams will only
publish papers as their ultimate goal, and will seriously consider cooperation with Taiwan's industry as one of the evaluation criteria. One of
the most important criteria is "problems given by the industry, and solved by the academia." My research team then chose to cooperate with
the listed IC company ELAN. We hope to integrate respective software and hardware advantage, and use this to develop "smart city tra c
ow solutions" that are not only applicable to Taiwan, but also export internationally. We expect to complete the following systems in four
years: (1) A sub-system that can directly execute edge computing at intersections to compute various parameters of tra c ow; (2) A sub-
system that can use computer vision technology to compute other tra c ow parameters between adjacent intersections; and (3) A sub-
system that can make use of the parameters obtained above to trigger a reinforcement learning process, and then dynamically adjust all
tra c signs in a certain range.
Publications
1. C. Y. Wang, H. Y. Mark Liao, P. Y. Chen, and J. W. Hsieh, 6. Nick C. Tang, Yen-Yu Lin, Ming-Fang Weng, and Hong-Yuan Brochure 2020
"Enriching Variety of Layer-wise Learning Information by Mark Liao, "Cross-Camera Knowledge Transfer for Multiview
Gradient Combination," IEEE International Conference on People Counting," IEEE Transactions on Image Processing ,
Computer Vision Workshop (ICCVW) "Low Power Computer volume 24, number 1, pages 80-93, January 2015.
Vision.'', October 2019.
7. Y. L. Chen, C. T. Hsu, and H. Y. Mark Liao, "Simultaneous
2. Jen-Chun Lin, Wen-Li Wei, Tyng-Luh Liu, C.-C. Jay Kuo, and H. Tensor Decomposition and Completion Using Factor Priors,"
Y. Mark Liao, "Tell Me Where is Still Blurry: Adversarial Blurred IEEE Transactions on Pattern Analysis and Machine Intelligence,
Region Mining and Refining," 27th ACM Multimedia Conference volume 36, number 3, pages 577-591, March 2014.
(long paper), Nice, France, October 2019.
8. M. F. Weng, Y. Y. Lin, Nick C Tang, and H. Y. Mark Liao,
3. Jen-Chun Lin, Wen-Li Wei, Tyng-Luh Liu, Yi-Hsuan Yang, "Visual Knowledge Transfer among Multiple Cameras for People
Hsin-Min Wang, Hsiao-Rong Tyan, and Hong-Yuan Mark Counting with Occlusion Handling," 20th ACM Multimedia
Liao, "Coherent Deep-Net Fusion To Classify Shots In Concert Conference (long paper), October 2012.
Videos," IEEE Transactions on Multimedia, volume 20, number
11, pages 3123-3136, November 2018. 9. C. H. Ling, Y. M. Liang, C. W. Lin, Y. S. Chen, and H. Y. Mark
Liao, "Human Object Inpainting Using Manifold Learningbased
4. Jen-Chun Lin, Wen-Li Wei, James Yang, Hsin-Min Wang, and Posture Sequence Estimation," IEEE Trans. on Image Processing,
Hong-Yuan Mark Liao, "Automatic Music Video Generation volume 20, number 11, pages 3124-3135, November 2011.
Based on Simultaneous Soundtrack Recommendation and Video
Editing," 25th ACM Multimedia Conference (long paper), San 10. Hsueh-Yi Sean Lin, Hong-Yuan Mark Liao, and Ja-Chen
Jose, USA, October 2017. Lin, "Visual Salience-Guided Mesh Decomposition," IEEE
Transactions on Multimedia, volume 9, number 1, pages P46-P57,
5. P. C. Wen, W. C. Cheng, Y. S. Wang, H. K. Khu, Nick C. January 2007.
Tang, and H. Y. Mark Liao, "Court Reconstruction for Camera
Calibration in Broadcast Basketball Videos," IEEE Transactions
on Visualization and Computer Graphics, volume 22, number 5,
pages 1517-1526, May 2016.
147
