Audio-visual query and retrieval: A system that uses dynamic programming and relevance feedback

A necessary capability for content-based retrieval is to support the paradi gm of query by example. In the past, there have been several attempts to us e low-level features for video retrieval. However, most of the approaches s upport queries using image sequences only. We present an algorithm for matc hing multimodal (audio-visual) patterns for the purpose of content-based vi deo retrieval The novel ability of our approach to use the information cont ent in multiple media coupled with a strong emphasis on temporal similarity differentiates it from the state of the art in content-based retrieval. At the core of the pattern matching scheme is a dynamic programming algorithm , which leads to a significant improvement in performance. Coupling the use of audio with video this algorithm can be applied to grouping of shots bas ed on audio-visual similarity. We also support relevance feedback. The user can provide feedback to the system, by choosing clips, which are closer to the user's desired target. The system then automatically adjusts the relat ive weights or relevance of the media and fetches different sets of target clips accordingly. It is our observation that a few iterations of such feed back are generally sufficient, for retrieving the desired video clips. (C) 2001 SPIE and IS&T.