With developments of semiconductor technology and powerful signal processing tools, there has been a proliferation of personal digital media devices such as digital cameras, music and digital video players. In parallel, storage media (both magnetic and optical) have become cheaper to manufacture and correspondingly their capacities have increased. This has spawned new applications such as Multimedia Information Systems, CAD/CAM, Geographical Information systems (GIS), medical imaging, time-series analysis (in stock markets and sensors), that periodically store large amounts of high-dimensional data that are later retrieved and processed.
Thus, large repositories of high dimensional data are central to a vast array of disciplines and applications, and the degree to which they can be exploited depends critically on the availability of efficient and smart tools for search and retrieval, analysis, and mining. This ongoing research project is primarily concerned with the problem of efficient, interactive, approximate and exact similarity search in high-dimensional data sets. Our methodology involves approaches whose origins lie in several disciplines beside classical data management, including optimization, information theory, pattern recognition and signal compression.
Research team includes:
To our Sponsors - Thank You!: This project is mainly supported by the National Science Foundation (NSF) under grant IIS-0329267; and in part by the University of California MICRO program, Applied Signal Technology, Inc., Dolby Laboratories Inc., and Qualcomm Inc.
Some of our high-dimensional datasets are available for download here....
A list of publications is available here....
Some presented talks are available:
- Fusion Coding of Correlated Sources, Fusion Coder Design, Shared Descriptions Fusion Coder, Predictive Fusion Coder
- Cluster Distance Bounding, Relevance Feedback Indexing