Face Occlusion Detection Based on SSD Algorithm

Abstract

In recent years, as an important technology in deep learning, target detection has been widely used in all aspects of life. Aiming at the problem of occlusions in face recognition, this paper adopts SSD (Single Shot MultiBox Detector) deep learning target detection algorithm to classify and locate face occlusions. The average precision of all categories (mAP) reached 95.46% through the self-built data set of 7 types of common face occlusion. Experiments show that this method can effectively detect the face occlusion, which provides a new idea for automatic intelligent face recognition and has a broad application prospect.