Technological advances over the last several years have made it possible to construct a large digital image and video libraries comprising tens of terabytes of online or nearly online data. As a result of the continued proliferation of image and video data, these libraries will grow significantly larger over the next few years. As a consequence, the demand for the capability to provide databases that can effectively support storage, search, retrieval and transmission of this kind of nontraditional data will grow significantly. This paper describes a project currently in progress at the IBM T.J. Watson Research Center that proposes to explore some of these challenges. In order to store, manage, and retrieve data from a large archive, we propose a progressive framework that reorganize the image and video data into a pyramid. Consequently, search operations can be performed in a hierarchical fashion and thus significantly reduce the total amount of data that need to be processed. Dramatic speedup has been achieved for many search operations such as template matching, texture feature extraction, and classification. We also developed an intuitive while expressive drag-and-drop query front-end to allow constructing of complex queries.
| Project home| Technical agenda| Publications| Contact|