Multimedia and Machine Learning Lab

The Multimedia and Machine Learning (MML) Lab, founded in Sept. 2009, is under the direction of Dr. Y.-C. Frank Wang. MML is dedicated to research topics of pattern recognition, machine learning, computer vision, and multimedia content analysis. In particular, we have wide interests in various research topics for multimedia applications.

MML is one of the research labs funded by the Research Center for Information Technology Innovation (CITI) of Academia Sinica, the leading national research institution in Taiwan. Several research projects in MML are also partially supported by the National Science Council, which is the highest government agency in Taiwan responsible for promoting the development of science and technology.

• 2012-05-06 Openings for Postdoc and RA Positions

  MML Lab is looking for postdoc researchers, research assistants, and doctoral students to join our exciting projects on pattern recognition, machine learning, computer vision, and multimedia content analysis. Candidates with strong academic background and research experience are encouraged to apply. Candidates should email their CVs with a complete publication list to Dr. Frank Wang.

• 2012-05-01 Congratulations! MML's work is accepted for IEEE Transactions on Knowledge and Data Engineering.

  Congratulations to Yi-Ren Yeh, whose paper on "Anomaly Detection via Online Over-Sampling Principal Component Analysis" is accepted for IEEE TKDE. This is a joint work with Prof. Yuh-Jye Lee at National Taiwan University Science & Technology.

• 2012-04-18 Congratulations! MML has 2 papers accepted for IEEE ICIP 2012.

  Congratulations to Wei-Te Li, Hui-Tang Chang, Hermes Lyu, and Fang-Lin He, whose papers on "Automatic Saliency Inspired Foreground Object Extraction from Videos" and "A Self-Learning Approach to Color Demosaicking via Support Vector Regression" are accepted for IEEE ICIP 2012.

• 2012-03-01 Congratulations! MML's work is accepted for IEEE CVPR 2012.

  Congratulations to Chih-Fan Chen and Chia-Po Wei, whose paper on "Low-Rank Matrix Recovery with Structural Incoherence for Robust Face Recognition" is accepted for IEEE CVPR 2012.

• 2012-02-29 Congratulations! MML has two papers accepted for IEEE ICME 2012.

  Congratulations to Min-Chun Yang, De-An Huang, and Chih-Yun Tsai, whose works on learning-based single image super-resolution and context-aware single image rain removal have been accepted for ICME 2012.

• 2012-02-14 Congratulations! MML's work is accepted for IEEE Transactions on Multimedia.

  Congratulations to Yi-Ren Yeh, Ting-Chu Lin, and Yung-Yu Chung. Their paper on "A Novel Multiple Kernel Learning Framework for Heterogeneous Feature Fusion and Variable Selection" has been accepted for IEEE Transactions on Multimedia, ICME 2011 Special Issue. The paper is expected to be published in June 2012.

• 2012-01-09 Congratulations! MML achievements are selected as the cover story of Engineering Science & Technology Bulletin, National Science Council.

  Congratulations to MML members Min-Chun Yang, Ting-Yao Hu, and Chang-Heng Wang, whose work on learning-based single image super-resolution has been selected as the cover story of Engineering Science & Technology Bulletin, NSC in Jan. 2012.

• 2011-08-24 Congratulations! MML members receive the First Place Award in "Taiwan Tech Trek 2011" of NSC.

  Congratulations to MML members Shannon Chen (Dept. CS, Northwestern University) and Michael Cheng (Dept. EE, UCLA), who participated in TTT 2011 of NSC and worked as summer interns at MML, won the first place in Engineering-I track of TTT 2011.

• 2011-06-29 Congratulations! MML's work has been accepted for ACM Multimedia 2011.

  Congratulations to Chih-Fan Chen. His paper on "Exploring Self-Similarity of Bag-of-Features for Image Classification" is accepted for ACM Multimedia 2011 as a short paper.

• 2011-04-22 Congratulations! MML has two papers selected as Top 15% paper at ICME 2011.

  Congratulations to Kuo-Chin Lien, Chia-Ling Chang, and Kuan-Chieh Chen, who have their papers accepted as Top 15% paper at ICME 2011. One paper is on automatic object extraction in single-concept videos, and the other is on automatic annotation of Web videos. ICME was revamped in late 2009 to elevate it to the next level, which now follows a double-blind review process and targets at an acceptance rate below 30% for main conference papers.

For research issues and interested applicants, please contact

• Dr. Y.-C. Frank Wang

• Email: ycwang

• Telephone: +886-2-2787-2368

For administrative issues, please contact

• Ms. Maja Lee

• Email: maja

• Telephone: +886-2-2787-2349

Research Center for Information Technology Innovation, Academia Sinica
No. 128, Sec. 2, Academia Rd. Nangang District, Taipei 11529, Taiwan

Dr. Yu-Chiang Frank Wang Email: ycwang Phone: 2787-2368 Research Interests: Pattern Recognition, Machine Learning, Computer Vision, Multimedia Signal Processing and content analysis

Dr. Yi-Ren Yeh Email: yryeh Phone: 2787-2335 Research Interests: Machine learning, data mining, numerical optimization, and network security

Dr. Chia-Po Wei Email: cpwei Phone: 2787-2395 Research Projects: Sparse Representation for Visual Classification

Chih-Fan Chen Email: ryanchen Phone: 2787-2331 Research Projects: Object recognition and localization using semantic features

Ting-Chu Lin Email: tingchulin Phone: 2787-2332 Research Projects: Heterogeneous feature fusion for video retrieval

Wei-Te Li Email: wtli Phone: 2787-2332 Research Projects: Human detection, segmentation, and tracking in images and videos

Chiao-Meng Huang Email: chairman Phone: 2787-2331 Research Projects: Recommendation algorithms for multimedia data

Chun-Hao Huang Email: paulchhuang Phone: 2787-2331 Research Projects: Transfer learning

Ming-Chun Yang Phd Candidate from NTU CSIE Research Projects: Spatio-temporal super-resolution using learning-based methods

Maja Lee Email: maja Phone: 2787-2349

• Hui-Tang Chang (EE, National Taiwan University, Class of 2012), Feb. 2011 - present
• Hermes Shing Lyu (EE, National Taiwan University, Class of 2012), Feb. 2011 - present
• De-An Huang (EE, National Taiwan University, Class of 2012), Feb. 2011 - present
• Chih-Yun Tsai (EE, National Taiwan University, Class of 2012), Feb. 2011 - present

• Ming-Hen Tsai, Jul. 2011 - Aug. 2011 (Ph.D. student in CS, Columbia)
• Kuo-Chin Lien, Aug. 2010 - Jul. 2011 (Ph.D. student in ECE, University of California, Santa Barbara)
• Yu-Wei Chao, Aug. 2010 - Jul. 2011 (Master student in ECE, University of Michigan, Ann Arbor)
• Ching-Hui Chen, Jun. 2010 - Jul. 2011 (Ph.D. student in ECE, University of Maryland)
• Chia-Ling Chang, Aug. 2010 - Jan. 2011 (Master student in ECE, University of Michigan, Ann Arbor)
• Wang-Chou Lu, Aug. 2009 - Apr. 2010 (Garmin Corp.)
• Chao-Chung Chu, Oct. 2009 - May. 2010 (MStar Semiconductor Inc.)
• Shao-Chuan Wang, Aug. 2009 - Jun. 2010 (Microsoft Corp.)
• Kuan-Chieh Chen, Feb. 2010 - Jun. 2010 (Ph.D. student in BME/ECE, CMU)
• Yung-Yu Chung, Nov. 2009 - Jul. 2010 (Ph.D. student in ECE, Iowa State University)
• Ting-Yao Hu (EE, National Taiwan University, 2011), Feb. 2010 - Jun. 2011
• Chang-Heng Wang (EE, National Taiwan University, 2011), Feb. 2010 - Jun. 2011
• Yu-Wen Chen (EE, National Taiwan University, 2010), Oct. 2009 - Feb. 2011
• Liang-Gang Huang (CSIE, National Taiwan University, 2010), Feb. 2010 - Jul. 2010
• Hsin-Yi Ho (CSIE, National Taiwan University, 2011), Oct. 2009 - Jul. 2010

C.-F. Chen, C.-P. Wei, and Y.-C. F. Wang, "Low-Rank Matrix Recovery with Structural Incoherence for Robust Face Recognition," IEEE International Conference on Computer Vision & Pattern Recognition (CVPR), 2012 (acceptance rate: 24%).(PDF)

Y.-R. Yeh, T.-C. Lin, Y.-Y. Chung, and Y.-C. F. Wang, "A Novel Multiple Kernel Learning Framework for Heterogeneous Feature Fusion and Variable Selection," To appear in IEEE Transactions on Multimedia. (PDF)

C.-F. Chen and Y.-C. F. Wang, "Exploring Self-Similarity of Bag-of-Features for Image Classification," ACM Multimedia, 2011. (PDF)(Poster)

K.-C. Lien and Y.-C. F. Wang, "Automatic Object Extraction in Single-Concept Videos," IEEE International Conference on Multimedia & Expo (ICME), 2011 (Accept as Top 15% paper; overall acceptance rate 30%) (PDF) Kuo-Chin Lien, Wei-Te Li and Yu-Chiang Frank Wang, "Unsupervised Foreground Object Extraction in Single-Concept Videos," submitted to IEEE Transactions on Multimedia Demo Video: (Tracking)(Detection)(Unsupervised).

M.-C. Yang, C.-H. Wang, T.-Y. Hu, and Y.-C. F. Wang, "Learning Context-Aware Sparse Representation for Single Image Super-Resolution," IEEE International Conference on Image Processing (ICIP), 2011. (PDF) M.-C. Yang, C.-T. Chu, and Y.-C. F. Wang, "Learning Sparse Image Representation with Support Vector Regression for Single Image Super-Resolution," IEEE International Conference on Image Processing (ICIP), 2010. (PDF)(code)

Y.-W. Chao, Y.-R. Yeh, Y.-W. Chen, Y.-J. Lee, and Y.-C. F. Wang, "Locality-constrained Group Sparse Representation for Robust Face Recognition,," IEEE International Conference on Image Processing (ICIP), 2011. (PDF)

S.-C. Wang and Y.-C. F. Wang, "A Multi-Scale Learning Framework for Visual Categorization," Asian Conference on Computer Vision (ACCV), Nov. 2010 (Oral paper. Acceptance rate: 29.5% overall, 4.7% oral). (PDF)

National Science Council

• Sparse-Coded Image and Video Object Extraction for Human Action Recognition, PI, 2010/08 - 2011/07.

• Compressive Sensing Techniques for Multimedia Research and Applications, co-PI, 2010/08 - 2011/07.

• Taiwan e-Learning and Digital Archives Program, co-PI, 2010/01 - 2011/12.

• A Super-Pixel based Automatic Learning Framework for Object Categorization, PI, 2009/11 - 2010/10.

Industrial Technology Research Institute

• Intelligent Video Surveillance Technology, PI, 2011/05 - 2011/12.

Chunghwa Telecom Co., Ltd.

• Online Network Behavior Anomaly Detection, PI, 2011/05 - 2012/04.

M.-C. Yang and Y.-C. F. Wang, "A Self-Learning Approach to Single Image Super-Resolution," IEEE Transactions on Multimedia, Under Revision.

Y.-R. Yeh and Y.-C. F. Wang, "A Rank-One Update Method for Least Squares Linear Discriminant Analysis with Concept Drift," Pattern Recognition, Under Revision.

Y.-J. Lee, Y.-R. Yeh, and Y.-C. F. Wang, "Anomaly Detection via Online Over-Sampling Principal Component Analysis," To appear in IEEE Transactions on Knowledge and Data Engineering.

Y.-R. Yeh, T.-C. Lin, Y.-Y. Chung, and Y.-C. F. Wang, "A Novel Multiple Kernel Learning Framework for Heterogeneous Feature Fusion and Variable Selection," IEEE Transactions on Multimedia, volume 14, number 3, pages TBD, June 2012.

Y.-C. F. Wang and D. Casasent, "New support vector-based design method for binary hierarchical classifiers for multi-class classification problems," Neural Networks, (21) 2-3, Mar. - Apr. 2008, pp. 502-510.  (PDF)

D. Casasent and Y.-C. Wang, "A hierarchical classifier using new support vector machine for automatic target recognition," Neural Networks, (18) 5-6, Jul. - Aug. 2005, pp. 541-548.  (PDF)

2012

C.-F. Chen, C.-P. Wei, and Y.-C. F. Wang, "Low-Rank Matrix Recovery with Structural Incoherence for Robust Face Recognition," IEEE International Conference on Computer Vision & Pattern Recognition (CVPR), 2012 (acceptance rate: 24%).(PDF)

M.-C. Yang*, D.-A. Huang*, C.-Y. Tsai, and Y.-C. F. Wang, "Self-Learning of Edge-Preserving Single Image Super-Resolution via Contourlet Transform," IEEE International Conference on Multimedia & Expo (ICME), 2012 (* indicates equal contribution; acceptance rate: 30%).

D.-A. Huang, L.-W. Kang, C.-Y. Tsai, M.-C. Yang, C.-W. Lin, and Y.-C. F. Wang, "Context-Aware Single Image Rain Removal," IEEE International Conference on Multimedia & Expo (ICME), 2012 (acceptance rate: 30%).

W.-T. Li, H.-T. Chang, H. S. Lyu, and Y.-C. F. Wang, "Automatic Saliency Inspired Foreground Object Extraction from Videos," IEEE International Conference on Image Processing (ICIP), 2012.

F.-L. He, Y.-C. F. Wang, and K.-L. Hua, "A Self-Learning Approach to Color Demosaicking via Support Vector Regression," IEEE International Conference on Image Processing (ICIP), 2012.

2011

C.-F. Chen and Y.-C. F. Wang, "Exploring Self-Similarity of Bag-of-Features for Image Classification," ACM Multimedia, 2011.(PDF)

M.-C. Yang, T.-Y. Hu*, C.-H. Wang*, and Y.-C. F. Wang, "Learning of Context-Aware Single Image Super-Resolution," IEEE Visual Communication and Image Processing, 2011 (invited paper; * indicates equal contribution).

T.-Y. Lin, Y.-Y. Lin, M.-F. Wang, Y.-C. F. Wang, Y.-F. Hsu, and H.-Y. M. Liao, "Cross Camera People Counting with Perspective Estimation and Occlusion Handling," IEEE International Workshop on Information Forensics and Security (WIFS), 2011 (acceptance rate: 32%).

Y.-W. Chao, Y.-R. Yeh, Y.-W. Chen, Y.-J. Lee, and Y.-C. F. Wang, "Locality-constrained Group Sparse Representation for Robust Face Recognition," IEEE International Conference on Image Processing (ICIP), 2011.(PDF)

M.-C. Yang, C.-H. Wang*, T.-Y. Hu*, and Y.-C. F. Wang, "Learning Context-Aware Sparse Representation for Single Image Super-Resolution," IEEE International Conference on Image Processing (ICIP), 2011. (* indicates equal contribution)(PDF)

C.-H. Chen, W.-H. Chung, and Y.-C. F. Wang, "Joint Source-Channel Coding Optimization with Packet Loss Resilience for Video Transmission," IEEE International Conference on Image Processing (ICIP), 2011.

C.-H. Chen, W.-H. Chung, and Y.-C. F. Wang, "Cross-Layer Design for Video Streaming with Dynamic Antenna Selection," IEEE International Conference on Image Processing (ICIP), 2011.

Y.-R. Yeh, Y.-Y. Chung, T.-C. Lin, and Y.-C. F. Wang, "Group Lasso Regularized Multiple Kernel Learning for Heterogeneous Feature Selection," IEEE International Joint Conference on Neural Networks (IJCNN), 2011.(PDF)

S.-W. Sun, Y.-C. F. Wang, Y.-L. Hung, C.-L. Chang, S.-S. Chen, H.-M. Wang, and H.-Y. M. Liao, "Automatic Annotation of Web Videos," IEEE International Conference on Multimedia & Expo, 2011 (Accept as Top 15% paper; overall acceptance rate 30%).(PDF)

K.-C. Lien and Y.-C. F. Wang, "Automatic Object Extraction in Single-Concept Videos," IEEE International Conference on Multimedia & Expo, 2011 (Accept as Top 15% paper; overall acceptance rate 30%).(PDF)

Y.-R. Yeh and Y.-C. F. Wang, "Least-Squares LDA via Rank-One Updates with Concept Drift," IEEE International Workshop on Statistical Signal Processing (SSP), 2011.(PDF)

2010

S.-C. Wang and Y.-C. F. Wang, "A Multi-Scale Learning Framework for Visual Categorization," Asian Conference on Computer Vision (ACCV), Nov. 2010 (Oral paper. Acceptance rate: 29.5% overall, 4.7% oral).(PDF)

M.-C. Yang, C.-T. Chu, and Y.-C. F. Wang, "Learning Sparse Image Representation with Support Vector Regression for Single-Image Super-Resolution," IEEE Int'l Conference on Image Processing, Sep. 2010. (PDF)(code)

W.-C. Lu, Y.-C. F. Wang, and C.-S. Chen, "Video Object Extraction by Dense Optical-Flow Trajectories," IEEE Int'l Conference on Advanced Video and Signal Based Surveillance, Aug. 2010.

S.-C. Wang and Y.-C. F. Wang, "Simultaneous Object Recognition and Localization in Image Collections," IEEE Int'l Conference on Advanced Video and Signal Based Surveillance, Aug. 2010.(PDF)

S.-C. Wang and Y.-C. F. Wang, "Segmentation-free object localization in image collections," IEEE Workshop on Visual Content Identification and Search, in conjunction with ICME, Jul. 2010.

Before 2010

Y.-C. F. Wang and D. Casasent, "A support vector hierarchical method for multi-class classification and rejection," IEEE Int'l Joint Conf. on Neural Networks, Jun. 2009. (PDF)

Y.-C. F. Wang and D. Casasent, "Soft-decision hierarchical classification using SVM-type classifiers," IEEE Int'l Joint Conf. on Neural Networks, Jun. 2008. 

Y.-C. F. Wang and D. Casasent, "New design method for a hierarchical SVM-based classifier," Proc. SPIE,Vol. 6764, Sep. 2007.

Y.-C. F. Wang and D. Casasent, "New weighted support vector k-means clustering for hierarchical multi-class classification," IEEE Int'l Joint Conf. on Neural Networks, Aug. 2007. 

Y.-C. F. Wang and D. Casasent, "A new SVM for scale, aspect, and depression angle tolerant IR object recognition," Proc. SPIE, Vol. 6574, Apr. 2007.

Y.-C. F. Wang and D. Casasent, "Hierarchical k-means clustering using new support vector machine for multi-class classification," IEEE Int'l Joint Conf. on Neural Networks, Jul. 2006.

Y.-C. Wang and D. Casasent, " New hierarchical SVM classifier for multi-class target recognition," Proc. SPIE, Vol. 6006, Oct. 2005, pp. 359-370.

Y.-C. Wang and D. Casasent, "A hierarchical classifier using new support vector machines," IEEE Int'l Conf. on Document Analysis and Recognition, Aug. 2005, pp. 851-855.

D. Casasent and Y.-C. Wang, "Automatic target recognition using new support vector machine," IEEE Int'l Joint Conf. on Neural Networks, Aug. 2005.

Y.-C. Wang and D. Casasent, "Multimodal biometric fusion using data quality information," Proc. SPIE, Vol. 5816, March 2005, pp. 329-338. 

D. Casasent and Y.-C. Wang, "New distortion-invariant SVRDM hierarchial classifier," Proc. SPIE, Vol. 5608, Oct. 2004, pp. 329-337.

Jul. 1, 2012	Asian Conference on Computer Visio (ACCV 2012)	Daejeon, Korea, Nov. 5-9, 2012
Nov. 4, 2012	IEEE Conf. on Computer Vision and Pattern Recognition (CVPR 2013)	Portland, USA, Jun. 23-28, 2013
Nov. 19, 2012	IEEE Int'l Conf. on Acoustics, Speech, and Signal Processing (ICASSP 2012)	Vancouver, Canada, May 26-31, 2013
Jan. 15, 2013	IEEE Int'l Conference on Multimedia and Expo (ICME 2013)	San Jose, USA, July 15-19, 2013

Submission Deadline	Special Issues	Journal
Apr. 1, 2011	New Image and Video Representations Based on Sparsity	EURASIP Journal on Advances in SIgnal Processing
Apr. 1, 2011	Image and Video Quality Improvement Techniques for Emerging Applications	EURASIP Journal on Advances in SIgnal Processing
Jun. 16, 2011	Learning Semantics from Multimedia Wed Resources	IEEE Transactions on Multimedia