Face straightening method based on tripartite confrontation generation network
作者： Li Minjun , Cheng Juntong , Jiang Yugang
DOI:
关键词:
摘要: The invention belongs to the technical field of computer image generation, and particularly relates a face straightening method based on tripartite confrontation generation network. According tothe method, two generators are introduced, namely contour generator facial feature which respectively used for correcting training generators, so that generated front better retains original features, is clear; discriminator D introduced confront with picture vivid; and, circular consistent loss accuracy (identity expression feature) retention improved. large-angle side can be straightened, information restored.
lens.org 参考文章(12)
Jan Kautz, Ming-Yu Liu, Xiaodong Yang, Sergey Tulyakov, System and method for content and motion controlled action video generation ,(2020)
Zhang Yunxuan, Li Cheng, Qian Chen, Wu Wenyan, Face reconstruction method, apparatus and device based on reconstruction network, medium and product ,(2018)
Kim Ji Won, Cha Moon Su, Kim Taek Soo, Method for learning cross-domain relations based on generative adversarial networks ,(2018)
Huang Jiabo, Zheng Weishi, Xie Xiaohua, Multi-image face alignment method and device based on generative adversarial network ,(2018)
Jiaxin Ma, Feng Zhou, Multi-poses Face Frontalization based on Pose Weighted GAN 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC). ,(2019) , 10.1109/ITNEC.2019.8729088
Luo Ping, Yan Junjie, Shen Yujun, Image generation method and device, equipment and medium ,(2019)
Kaufhold John Patrick, Sleeman Jennifer Alexander, Systems and methods for deep model translation generation ,(2020)
Yu Jun, Gao Fei, Sun Kejia, Face image enhancement method based on generative adversarial network ,(2018)
Huang Guoheng, Cheng Lianglun, Zhao Zhishun, A method for generating a multi-angle face image based on a generative adversarial network model ,(2019)