Discovering similarities for content-based recommendation and browsing in multimedia collections

Taras Lehinevych, Nikolaos Kokkinis-Ntrenis, Giorgos Siantikos, A Seza Dogruoz, Theodoros Giannakopoulos, and Stasinos Konstantopoulos
Signal-Image Technology and Internet-Based Systems (SITIS), 2014 Tenth International Conference on. IEEE, 2014

The purpose of the research described in this paper is to examine the existence of correlation between low level audio, visual and textual features and movie content similarity. In order to focus on a well defined and controlled case, we have built a small dataset of movie scenes from three sequel movies. In addition, manual annotations have led to a ground-truth similarity matrix between the adopted scenes. Then, three similarity matrices (one for each medium) have been computed based on Gaussian Mixture Models (audio and visual) and Latent Semantic Indexing (text). We have evaluated the automatically extracted similarities along with two simple fusion approaches and results indicate that the low-level features can lead to an accurate representation of the movie content.

BibTeX, IEEE, Paper (PDF, 673 KB)

@inproceedings{lehinevych2014discovering,
  title={Discovering similarities for content-based recommendation and browsing in multimedia collections},
  author={Lehinevych, Taras and Kokkinis-Ntrenis, Nikolaos and Siantikos, Giorgos and Dogru{\"o}z, A Seza and Giannakopoulos, Theodoros and Konstantopoulos, Stasinos},
  booktitle={Signal-Image Technology and Internet-Based Systems (SITIS), 2014 Tenth International Conference on},
  pages={237--243},
  year={2014},
  organization={IEEE}
}