Face Retrieval in Videos using Face Quality Assessment and Convolution Neural Networks

With the large amount of videos produced every day, Content-Based Video Retrieval (CBVR) has become a necessity by describing each video with a compact and significant signature in order to efficiently retrieve the desired video from a large collection. In this work, we present a CBVR system applied on face recognition based on key frames. The first step in this system consists of extracting key frames from videos using Face Quality Assessment (FQA) and Convolution Neural Networks (CNN). Starting by generating face quality scores for each face image using three face feature descriptors (Gabor, Local Binary Pattern (LBP) and Histogram of Oriented Gradient (HoG)). Then, we train a Convolution Neural Network (CNN) in a supervised manner in order to select frames having the best face quality. Experiments on several datasets has shown that the proposed “DeepFQA” method gives promising results in terms of accuracy and precision/recall curve.