Multi-modal analysis for person type classification in news video
Jun Yang, Alexander G. Hauptmann · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2005
Classifying the identities of people appearing in broadcast news video into anchor, reporter, or news subject is an important topic in high-level video analysis, which remains as a missing piece in the existing research. Given the visual resemblance of different types of people, this work explores multi-modal features derived from a variety of evidences, including the speech identity, transcript clues, temporal video structure, named entities, and face information. A Support Vector Machine (SVM) model is trained on manually-classified people to combine the multitude of features to predict the types of people who are giving monologue-style speeches in news videos. Experiments conducted on ABC World News Tonight video have demonstrated that this approach can achieve over 93% accuracy on classifying person types. The contributions of different categories of features have been compared, which shows that the relatively understudied features such as speech identities and video temporal structure are very effective in this task.