High-Dimensional and Low-Quality Visual Information Processing: From Structured Sensing and Understanding

ISBN-13: 9783662525630 / Angielski / Miękka / 2016 / 99 str.

Yue Deng

Opis

After exploring the inherent structures of the visual data, it proposes a number of computational models covering an extensive range of mathematical topics, including compressive sensing, graph theory, probabilistic learning and information theory.

Spis treści

Introduction.- Sparse Structure for Visual Signal Sensing.- Graph Structure for Visual Signal Sensing.- Discriminative Structure for Visual Signal Understanding.- Information Theoretic Structure for Visual Signal Understanding.- Conclusions.

O autorze

Bio Yue Deng received the B.E. degree (Hons.) in automatic control from Southeast University, Nanjing,
China, in 2008 and Ph.D. degree (Hons.) in control science from the Department of Automation, Tsinghua University, Beijing, China, in 2013. He was a Visiting Scholar with the School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, from 2010 to 2011. He is currently an Associate Professor with the School of Electronic Science and Engineering, Nanjing University, Nanjing, China.
Zhang,"Graph
Dr. Deng’s current research interests include computer vision, machine learning and computational biology. He is the first author of more than 10 papers published on the worldwide reputable journals and conferences. He was a recipient of the Microsoft global fellowship for young computer scientist in 2010, National Ph.D Scholarship in 2012, best control award of International Air Robotic competition in 2012. His dissertation was nominated as the outstanding Ph.D thesis by Tsinghua University in 2013.

Honors:
• Outstanding Ph.D thesis award of Tsinghua University, 2013
• Best Control Award of International Aerial Robotic Competition, 2012
• National Ph.D Fellowship, 2012
• First Class Scholarship of Tsinghua University, 2012
• Microsoft Global Fellowship for young Computer Scientist, 2010.
• First class award in Tsinghua-UC, Berkeley Global Technical Challenge Competition, 2010.
• Rockwell Global Scholarship, 2006.

Publications:
• Yue Deng, Q. Dai, R. Liu, Z. Zhang and S.Hu, "Low-Rank Structure Learning via Non-convex Heuristic Recovery", IEEE Transactions on Neural Network and Learning Systems, March, Pages 383-396, 2013 (Spotlight Paper)
• Yue Deng, Y. Liu, Q. Dai, Z. Zhang and Y. Wang, "Noisy Depth Maps Fusion for Multiview Stereo Via Matrix Completion", IEEE Journal of Selected Topics in Signal Processing, (IEEE J-STSP) Sep. Issue, 2012.
• Yue Deng, Q. Dai and Z. Zhang, "Graph Laplace for Partially Occluded Face Completion and Recognition", IEEE Transactions on Image Processing, (IEEE TIP) vol.20(8), 2329-2338, 2011.
• Yue Deng, Y. Zhao, Y.Liu and Q. Dai "Differences Help Recognition: A Probabilistic Interpretation", Plos ONE, 2013.
• Yue Deng, Q. Dai, R. Wang and Z. Zhang, "Commute Time Guided Transformation for Feature Extraction", Computer Vision and Image Understanding Vol. 116, Issue 4, April, Pages 473-483,2012.
• Yue Deng, and Q. Dai, "An Overview of Computational Sparse Models and Their Applications in Artificial Intelligence", Artificial Intelligence Book, Springer, 2012. (book chapter)
• Yue Deng, Y. Qian, Y. Li and Q. Dai, "Visual Words Assignment on A graph via Minimal Information Loss", British Machine Vision Conference (BMVC), 2012.
• Yue Deng, D. Li, X. Xie, K. Lam and Q. Dai, "Partially Occluded Face Completion and Recognition", IEEE International Conference on Image Processing (ICIP), 2009.
• Yue Deng, Q. Dai and Z. Zhang, "Feature extraction using randomwalks", IEEE Youth Conference on Information, Computing and Telecommunication, 2009. YC-ICT’09. IEEE, 2009.
• Genzhi Ye, Yebin Liu, Yue Deng, Nils Hasler, Xiangyang Ji, Qionghai Dai, Christian Theobalt, Free-viewpoint Video of Human Actors using Multiple Handheld Kinects, in IEEE Trans. Cybernetics, VOL. 43, NO. 5, 2013.

Blurb

This thesis primarily focuses on how to carry out intelligent sensing and understand the high-dimensional and low-quality visual information. After exploring the inherent structures of the visual data, it proposes a number of computational models covering an extensive range of mathematical topics, including compressive sensing, graph theory, probabilistic learning and information theory. These computational models are also applied to address a number of real-world problems including biometric recognition, stereo signal reconstruction, natural scene parsing, and SAR image processing.