Recent research trends in Content-based Video Retrieval have shown topic models as an effective tool to dealwith the semantic gap challenge. In this scenario, this paper has a dual target: (1) it is aimed at studying howthe use of different topic models (pLSA, LDA and FSTM) affects video retrieval performance; (2) a novel incrementaltopic model (IpLSA) is presented in order to cope with incremental scenarios in an effective and efficientway. A comprehensive comparison among these four topic models using two different retrieval systems and tworeference benchmarking video databases is provided. Experiments revealed that pLSA is the best model in sparseconditions, LDA tend to outperform the rest of the models in a dense space and IpLSA is able to work properly inboth cases.
展开▼
