Video Data Mining

Data mining techniques that are successful in text and transaction data may not simply apply to image data that are non-structured. It is not a trivial task to discover meaningful visual patterns in image databases, because the spatial dependency and content variations in the visual data greatly challenge most existing methods.

We propose a novel and effective multilevel approach to cope with these difficulties for mining spatial co-location visual patterns. Specifically, the novelty of this work lies in the following components: (1) an efficient pattern discovery and summarization method that handles spatial dependency; and (2) an effective multilevel probabilistic method to tame the content uncertainties of visual patterns. This new approach learns a hierarchical generative model of the image database in an unsupervised fashion, and can be applied to various applications.



Demo Sequences of On-line Data Mining
[These videos are for the purpose of demonstrating the technology only. Any reproduction or propagation of these video is explicitly forbidden and the privacy of the persons in the video should be protected to the full extent.]
 
A dance group
A boy in a pool
A mom with the daughter
Note: [top-left] results of on-line video data mining that identifies video context of the target (e.g., the headin this case). [top-right] robust information integration that combines all the predictions from the video context.
A boy in a room
A girl and her dady in woods
Note: [bottom-left] comparison with a dedicated head tracker. [bottom-right] our result where the yellow bounding box highlights the output the tracked target.
The major subject of each sequence is a small kid, and the tracking of the head of the kid is of great interest. These sequences exhibist many challenges such as occlusion, background clutter, appearance changes, etc. In addition, these video are recorded by amateurs with a hand-held camera. The tracking task here is very difficult. We developed a method that introduces on-line data mining to visual tracking, which automatically discovers the visual context of the target and the use of the context make the tracking quite stable and robust (for details please see our papers).


Publication:
  1. Ming Yang, Gang Hua and Ying Wu, "Context-Aware Visual Tracking", IEEE Trans. on Pattern Analysis and Machine Intelligence, 2008

  2. Ming Yang, Ying Wu and Shihong Lao, "Intelligent Collaborative Tracking by Mining Auxiliary Objects", in Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR 06), New York City, NY, June 17-22, 2006.  [PDF]

  3.  
  4. Junsong Yuan, Ying Wu and Ming Yang, "Discovery of Collocation patterns: from Visual Words to Visual Phrases", in Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR'07), Minneapolis, MN, June 2007 

  5.  
  6. Junsong Yuan, Ming Yang and Ying Wu, "From Frequent Itemsets To Semantically Meaningful Visual Patterns", in Proc. ACM Int'l Conf. on Knowledge Discovery and Data Mining (SIGKDD'07), San Jose, CA, August 2007 

  7.  
  8. Ming Yang, Ying Wu and Shihong Lao, "Mining Auxiliary Objects for Tracking by Multibody Grouping", in Proc. IEEE Int'l Conf. on Image Processing (ICIP'07), San Antonio, TX, Sept. 2007 

  9.  
  10. Junsong Yuan, Zhu Li, Yun Fu, Ying Wu and Thomas S. Huang, "Common Spatial Pattern Discovery by Efficient Candidate Pruning", in Proc. IEEE Int'l Conf. on Image Processing (ICIP'07), San Antonio, TX, Sept. 2007 

  11.  
  12. Fan Jiang, Ying Wu and Aggelos K. Katsaggelos, "Abnormal Event Detection From Surveillance Video By Dynamic Hierarchical Clustering", in Proc. IEEE Int'l Conf. on Image Processing (ICIP'07), San Antonio, TX, Sept. 2007 


Back to Video Research


Updated 07/2006. Copyright © 2003-2006 Ying Wu