# Face Entity Recognition In The Film 'Notting Hill'

740 Words3 Pages
III. Face Entity Recognition: Based on the scene segmentation results, we can compute the face occurrence matrix O face =[oik ] m×n on each scene, where m is the number of faces, and n is the number of scenes. Here Oface has the same size with Oname because the number of face clusters is set the same as the number of distinct names in the script. Finally, the face affinity network which is represented by a matrix R face =[rij ]m×n. Table II demonstrates the face affinity matrix of some face clusters derived from the video of the film “Notting Hill”. All the values are normalized into the interval [0,1]. IV.Graph Matching The generated affinity matrix of name and face have some statistic properties of the characters which are relatively stable and insensitive to the noises, such as character A has more affinities with character B than C, character D has never co-occurred with character A, etc. We assume that while the absolute quantitative affinity values are changeable, the relative affinity relationships between characters (e.g., A is closer to B than to C) and the qualitative affinity values (e.g., whether D has co-occurred with A) usually remain unchanged. Therefore propose scheme represent the character co-occurrence in rank order. We denote the original affinity matrix as R={ rij }N×N , where N is the number of characters. First we look at the cells along the main diagonal (e.g., A co-occur with A, B co-occurs with B). We rank the diagonal affinity