Design And Implementation Of Face Effect System Based On Deep Learning

With the development of mobile Internet and the improvement of people’s aesthetic ability,various image effect systems have been widely used.The primary problem of face effect system is that in low-light conditions face effects often disappear or the effect becomes worse due to the failure to detect face.Face Detection is the first step to achieve the face effect.and the problem that the accuracy of face detection model decreases dramatically in low-light conditions is also a hot research topic at home and abroad.In order to improve the applicability of face detection in complex conditions and optimize the effects of image effects systems,the paper analyzes the problems and challenges encountered in face detection in low-light conditions,proposes a face detection model based on domain adaptation(DA),and studies the implementation of face cartoon effect based on generative adversarial networks(GAN),designs and implements a face effect system suitable for low-light conditions.The main innovations are as follows:In order to solve the problem that the accuracy of face detection decreases in low-light condition,this paper proposes an improved face detection model based on pixel-level domain adaptation,which aligns the image-level features and pixel-level features of unlabeled low-light face images and labeled normal-light face images through the unsupervised domain adaptation network framework.A multi-scale feature alignment module is added to solve the problem that is more difficult to detect tiny face.The detection accuracy(AP)of the model on the public low-light face dataset DARK FACE is 6.47 higher than baseline,and it also achieves better results compared with DSFD,Pyramid Box and other mainstream detectors.A face effect system is designed and implemented in this paper.The whole system consists of a mobile terminal,a server,and a database.The mobile terminal is used for realizing the interactive operation of users,obtain images,and display special effects,etc.The server is used for realizing data transmission and rendering special effects.The database is used to realize data storage and management.This system effectively reduces the influence of lighting conditions on the special effects of the face,and achieves more realistic and natural face effects,the solution proposed in this paper is practiced and demonstrated.