Under the direction of Zhongfei (Mark) Zhang, the Multimedia Research Lab conducts research in the areas of multimedia information indexing and retrieval, data mining and knowledge discovery, computer vision and image understanding, pattern recognition, bioinformatics and machine learning. Zhang, who holds a PhD in computer science from the University of Massachusetts at Amherst, joined Binghamton’s Computer Science Department in 1999.
He and his team have developed an automatic image annotation and retrieval method based on a probabilistic semantic model in which the visual features and textual words are connected via a hidden layer to constitute the concepts to be discovered automatically to exploit the synergy between the two modalities. This technology is less expensive and less limited than traditional manual annotation.
Zhang has also developed data-mining technologies that perform relational clustering analysis in a unified manner. The model can identify cluster structures consisting of not only each type of data object but also the interaction patterns between different types of objects. The novel approach can make sense and use of multi-type relational data by clustering multi-types of objects simultaneously. The power of such tools lies in their ability to reveal local and global groups and structures. One can apply algorithms to discover connections in shopping patterns, social networking, homeland security, gene patterns, epidemiology and numerous other fields.
In addition, Zhang’s group developed a method of detecting shadows and other time-specific objects in image databases. This work has application beyond aerial surveillance analysis, for example to the construction of pilot training environments, 3-D scientific reconstructions and battlefield planning and operations.
- System and Method for Image Annotation and Multi-Modal Image Retrieval Using Probabilistic Semantic Models (RB-209)
- Soft Correspondence Ensemble Clustering (RB-239)
- Rational Clustering, Multi-Type Relational data, Collective Factorization On Related Matrices (RB-240)
- Hierarchical Static Shadow Detection For Color Aerial Images (RB-176)
- Fast Automatic Detectionof Independent Motion From Compressed Surveillance Video (RB-142)