Never ending face recognition for the LoD Cloud

Julien Subercaze (1) , Christophe Gravier (1)

1 LT2C - Laboratoire Telecom Claude Chappe

In this paper, we present FoP, a never ending face recognition learner on the Web. The first iteration of the face recognition model pro- pelling FoP was trained using Freebase data about politicians and their pictures. FoP is a never-ending system: when a new face is recognized, the learned model is updated ac- cordingly. At this step, FoP is also giving data in return to the LoD cloud that fed him in the first place: it leverages Linked Data on people recognized in these pictures, and on which rectangle area. Provided that both machine learning models and the LoD cloud are continuously struggling for more data for refinement processes, we demonstrate how to take advantage of their respective advantages to publish and to link more data on multimedia fragments on the Web.

Domaines: Web

Julien Subercaze, Christophe Gravier. Never ending face recognition for the LoD Cloud. 2013. ⟨hal-00994431v1⟩