CONF Le_MEDIAEVAL_2015/IDIAP EUMSSI team at the MediaEval Person Discovery Challenge Le, Nam Wu, Di Meignier, Sylvain Odobez, Jean-Marc face clustering person discovery Speaker Diarization EXTERNAL https://publications.idiap.ch/attachments/papers/2015/Le_MEDIAEVAL_2015.pdf PUBLIC Working Notes Proceedings of the MediaEval 2015 Workshop Wurzen, Germany 2015 1613-0073 http://ceur-ws.org/Vol-1436/Paper71.pdf URL We present the results of the EUMSSI team's participation in the Multimodal Person Discovery task at the MediaEval challenge 2015. The goal is to identify all people who simultaneously appear and speak in a video corpus, which implicitly involves both audio stream and visual stream. We emphasize on improving each modality separately and benchmarking them to analyze their pros and cons.