Automatic extraction of face tracks is a key component of systems that analyse people in audio-visual content such as TV programs and movies. Due to the lack of properly annotated content of this type, popular algorithms for extracting face tracks have not been fully assessed in the literature. We introduce and make publicly available a new dataset, based… (More)
Accurate detection of facial landmarks is very important for many applications like face recognition or analysis. In this paper we describe an efficient detector of facial landmarks based on a cascade of boosted regressors of arbitrary number of levels. We define as many regressors as landmarks and we train them separately. We describe how the training is… (More)
Accounts are often shared by multiple users, eachof them having different item consumption and temporal habits. Identifying of the active user can lead to improvements ina variety of services by switching from account personalizedservices to user personalized services. To do so, we developa topic model extending the Latent Dirichlet Allocation usinga hidden… (More)
This paper describes a classification experience that was performed on the stream metadata collected by the Singing-fish search engine. Using the categorization created by the Singingfish Content Team, we have tried to assess the performances achievable by a simple na¨ıve Bayes classifier.