MAAS: Multi-modal Assignation for Active Speaker Detection
       Juan Leon Alcazar1 , Fabian Caba Heilbron2 , Ali K. Thabet1 & Bernard Ghanem1
             1
               King Abdullah University of Science and Technology (KAUST), 2 Adobe Research
     jc.leon@uniandes.edu.co, caba@adobe.com, ali.thabet@kaust.edu.sa, bernard.ghanem@kaust.edu.sa

             Abstract                    a) Individual Analysis       b) Holistic Assignation

  Active speaker detection requires a mindful integr ...