Analysis on click-through data from a very large search engine log shows that users are usually interested in the top-ranked portion of returned search results. Therefore, it is crucial for search engines to achieve high accuracy on the topranked documents. While many methods exist for boosting video search performance, they either pay less attention to the above factor or encounter difficulties in practical applications. In this paper, we present a flexible and effective re-ranking method, called CR-Re-ranking, to improve the retrieval effectiveness. To offer high accuracy on the top-ranked results, CRRe-ranking employs a cross-reference (CR) strategy to fuse multimodal cues. Specifically, multimodal features are first utilized separately to re-rank the initial returned results at the cluster level, and then all the ranked clusters from different modalities are cooperatively used to infer the shots with high relevance. Experimental results show that the search quality, especially on the top-ranked results, is improved significantly.
[1]
Monika Henzinger,et al.
Analysis of a very large web search engine query log
,
1999,
SIGF.
[2]
Thorsten Joachims,et al.
Optimizing search engines using clickthrough data
,
2002,
KDD.
[3]
Jun Yang,et al.
CMU Informedia's TRECVID 2005 Skirmishes
,
2005,
TRECVID.
[4]
Tsinghua University at TRECVID 2005
,
2005,
TRECVID.
[5]
G. P. Nguyen,et al.
The MediaMill TRECVID 2005 Semantic Video Search Engine (Draft Version).
,
2005
.
[6]
Tie-Yan Liu,et al.
Adapting ranking SVM to document retrieval
,
2006,
SIGIR.
[7]
Dong Xu,et al.
Columbia University TRECVID-2006 Video Search and High-Level Feature Extraction
,
2006,
TRECVID.
[8]
Nicu Sebe,et al.
Content-based multimedia information retrieval: State of the art and challenges
,
2006,
TOMCCAP.