A new multimedia information data mining method

Authors:
Jin Longcun;Wan Wanggen;Cui Bin;Yu Xiaoqing;Xu Hongwei
Affiliations:
Shanghai University, Shanghai, China;Shanghai University, Shanghai, China;Shanghai University, Shanghai, China;Shanghai University, Shanghai, China;Shanghai University, Shanghai, China
Venue:
Proceedings of the first ACM/SIGEVO Summit on Genetic and Evolutionary Computation
Year:
2009

Citing 8
Cited 1

Learning Patterns of Activity Using Real-Time Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
An integrated baseball digest system using maximum entropy method

Proceedings of the tenth ACM international conference on Multimedia
CLOSET+: searching for the best strategies for mining frequent closed itemsets

Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining
Bayesian hierarchical clustering

ICML '05 Proceedings of the 22nd international conference on Machine learning
Enhanced max margin learning on multimodal data mining in a multimedia database

Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining
Motion-based background subtraction using adaptive kernel density estimation

CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Editorial: Introduction to the Special Issue on Multimedia Data Mining

IEEE Transactions on Multimedia
Automatic soccer video analysis and summarization

IEEE Transactions on Image Processing

Intrinsic Bayesian model for high-dimensional unsupervised reduction

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we proposed an annotated multimedia information data mining method. We present a Bayesian hierarchical framework model for mining objects in multimedia data. The Multimedia can switch between different shots, the unknown objects can leave or enter the scene at multiple times, and the background can be clustered. The proposed framework model consists of annotation part and Bayesian hierarchical mining part. This algorithm has several advantages over traditional distance-based agglomerative mining algorithms. Bayesian hierarchical hypothesis testing is used to decide which merges are advantageous and to output the recommended depth of the tree. The framework model can be interpreted as a novel fast bottom-up approximate inference method for a process mixture model. We describe procedures for learning the model hyperparameters, computing the predictive distribution, and extensions to the framework model. Experimental results on virtual reality multimedia data sets demonstrate useful properties of the framework model.