Semantic Video Retrieval Using Audio Analysis
CIVR '02 Proceedings of the International Conference on Image and Video Retrieval
| Hi-index | 0.03 |
This paper addresses the dual problem of audio coding for compression and retrieval. Because of the sheer volume of multimedia data, much research effort has focused on coding techniques for the compression of this data. However, being able to store large amounts of data in a compact fashion has limited application if there is no meaningful way to access and interact with them. Traditionally, these two issues have been treated separately resulting in many problems for multimedia data management. The authors have proposed a solution to these problems for audio by developing a compact, perceptually based, structured audio representation that provides explicit support for content based retrieval, browsing and other interactions. In addition, the representation offers potential for compressed data storage. This paper outlines a coding scheme based on the representation and reports the compression results.