Lexing Xie

Keywords: Multimedia Analysis, Data Mining, Machine Learning

Research Staff Member, IBM T J Watson Research Center, Hawthorne, NY

Josef Raviv postdoc fellow, IBM T J Watson Research Center
PhD Student @ DVMM Group, Dept. of Electrical Engineering, Columbia University

Personal homepage: http://lexingxie.info
Email: xlx at {ee dot columbia dot edu, us dot ibm dot com}
Quick navigation:
research interests | recent publications | past projects | bio | resume


Research Interests

Problems relating to the mining, analysis and summarization of multimedia data. Relationships between vision, audition, natural language and comprehension. Novel techniques in signal processing, machine learning, and statistical pattern recognition. Applications to the information extraction and utilization from large multimedia corpora.


Recent Publications (full list)
Topic Type
visual recognition and annotation
machine learning
book chapter
event detection
refereed conference
media and communities
pattern mining
technical report
video search and retrieval
notebook or white papers
summarization and visualization
conference demo
media security
L. Xie, R. Yan, J. Yang, Multi-concept Learning with Large-scale Multimedia Lexicon. Proc. IEEE International Conference on Image Processing (ICIP 2008), San Diego, California, USA, October 2008. (PDF|POSTER)
J. He, S.-F. Chang, L. Xie, Fast Kernel Learning for Spatial Pyramid Matching. Proc. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR 2008), Anchorage, Alaska, USA, June 2008. (ABSTRACT|PDF)
L. Xie, H. Sundaram, M. Campbell, Event Mining in Multimedia Streams. Proceedings of the IEEE, v96, no4, pages 623-647, April 2008 (ABSTRACT|PDF)
L. Xie, R. Yan, Extracting Semantics from Multimedia Content: Challenges and Solutions. Book Chapter, in Multimedia Content Analysis:Theory and Applications, A. Divakaran Ed., Springer, 2008. (BOOKLINK|PDF)
L. Liu, L. Xie, D. J. Lan, P. N. Liu, R. Moulic, SpotSmart: A Mobile Shopping Assistant. ACM Mobicom 2008 Demo, San Franscisco, USA, Sept 2008
A. Natsev, A. Haubold, J Tesic, L. Xie, R. Yan, Semantic Concept-based Query Expansion and Re-ranking for Multimedia Retrieval. Proc. ACM Multimedia (ACM-MM 2007), Augsburg, Germany, October 2007. (ABSTRACT | PDF)
A. Zunjarwad, H. Sundaram, L. Xie, Contextual Wisdom: Social Relations and Correlations for Multimedia Event Annotation. Proc. ACM Multimedia (ACM-MM 2007), Augsburg, Germany, October 2007. (ABSTRACT | PDF)
L. Xie, A. Natsev, J. Tesic, Dynamic Multimodal Fusion in Video Search. Proc. IEEE Intl. Conf. on Multimedia and Expo (ICME) 2007, Beijing, China, July 2007. (ABSTRACT |PDF|SLIDES)
B. Shevade, H. Sundaram, L. Xie, Modeling Personal and Social Network Context for Event Annotation in Images. Proc. ACM/IEEE Joint Conf. on Digital Libraries (JCDL) 2007, June 2007, Vancouver, Canada. (ABSTRACT|PDF) [best student paper award]
Projects (2000-2005)
DotMuse -- Discovery Of Temporal structure from MUltimedia SEquences: This project deals with the problem of automatically discover repetitive structure from multimedia sequences using unsupervised learning. Further topics include feature selection, multi-modal integration, sparse event handling, etc.
Also see: [slides PDF] [description@dvmm]
Semi-supervised Learning of Patch Based Appearance Models. This work is concerned with learning object appearance from image patches under a semi-supervised scenario. Soccer! This project is concerned with the automatic identification of the important semantic states in domain-specific video using domain knowledge and suervised learning techniques, using soccer video as test data set. [@dvmm]
BenchSari: Benchmarking of SARI image watermarking system. (past project as of Nov. 2000) VisGenie: A generic visualization system for audio-visual stream and various metadata. Now celebrating its own release in 2005 at the new website, credits to its lead architecht Yong Wang.