DeepFaceEditing: Deep Face Generation and Editing with Disentangled Geometry and Appearance Control

doi:10.1145/3450626.3459760

	DeepFaceEditing: Deep Face Generation and Editing with Disentangled Geometry and Appearance Control
	Chen, Shu-Yu 1; Liu, Feng-Lin 1; Lai, Yu-Kun 2; Rosin, Paul L.2; Li, Chunpeng 1; Fu, Hongbo 3; Gao, Lin 1
刊名	ACM TRANSACTIONS ON GRAPHICS
	2021-08-01
卷号	40 期号:4 页码:15
关键词	Deep image generation face editing image disentangling sketch-based interfaces
ISSN号	0730-0301
DOI	10.1145/3450626.3459760
英文摘要	Recent facial image synthesis methods have been mainly based on conditional generative models. Sketch-based conditions can effectively describe the geometry of faces, including the contours of facial components, hair structures, as well as salient edges (e.g., wrinkles) on face surfaces but lack effective control of appearance, which is influenced by color, material, lighting condition, etc. To have more control of generated results, one possible approach is to apply existing disentangling works to disentangle face images into geometry and appearance representations. However, existing disentangling methods are not optimized for human face editing, and cannot achieve fine control of facial details such as wrinkles. To address this issue, we propose DeepFaceEditing, a structured disentanglement framework specifically designed for face images to support face generation and editing with disentangled control of geometry and appearance. We adopt a local-to-global approach to incorporate the face domain knowledge: local component images are decomposed into geometry and appearance representations, which are fused consistently using a global fusion module to improve generation quality. We exploit sketches to assist in extracting a better geometry representation, which also supports intuitive geometry editing via sketching. The resulting method can either extract the geometry and appearance representations from face images, or directly extract the geometry representation from face sketches. Such representations allow users to easily edit and synthesize face images, with decoupled control of their geometry and appearance. Both qualitative and quantitative evaluations show the superior detail and appearance control abilities of our method compared to state-of-the-art methods.
资助项目	National Natural Science Foundation of China[61872440] ; National Natural Science Foundation of China[62061136007] ; Science and Technology Service Network Initiative of the Chinese Academy of Sciences[KFJ-STS-ZDTP-070] ; Science and Technology Service Network Initiative of the Chinese Academy of Sciences[KFJ-STS-QYZD-129] ; Science and Technology Service Network Initiative of the Chinese Academy of Sciences[KFJ-STS-QYZD-2021-11-001] ; Royal Society Newton Advanced Fellowship[NAF\R2\192151] ; Youth Innovation Promotion Association CAS ; Beijing Program for International S&T Cooperation Project[Z191100001619003] ; HK-SAR RGC General Research Fund[11212119] ; City University of Hong Kong (SCM ACIM Collaborative Research Fellowship)
WOS研究方向	Computer Science
语种	英语
出版者	ASSOC COMPUTING MACHINERY
WOS记录号	WOS:000674930900056
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/17284]
专题	中国科学院计算技术研究所
通讯作者	Chen, Shu-Yu
作者单位	1.Chinese Acad Sci, Beijing Key Lab Mobile Comp & Pervas Device, Inst Comp Technol, Beijing, Peoples R China 2.Cardiff Univ, Sch Comp Sci & Informat, Cardiff, Wales 3.City Univ Hong Kong, Sch Creat Media, Hong Kong, Peoples R China
推荐引用方式 GB/T 7714	Chen, Shu-Yu,Liu, Feng-Lin,Lai, Yu-Kun,et al. DeepFaceEditing: Deep Face Generation and Editing with Disentangled Geometry and Appearance Control[J]. ACM TRANSACTIONS ON GRAPHICS,2021,40(4):15.
APA	Chen, Shu-Yu.,Liu, Feng-Lin.,Lai, Yu-Kun.,Rosin, Paul L..,Li, Chunpeng.,...&Gao, Lin.(2021).DeepFaceEditing: Deep Face Generation and Editing with Disentangled Geometry and Appearance Control.ACM TRANSACTIONS ON GRAPHICS,40(4),15.
MLA	Chen, Shu-Yu,et al."DeepFaceEditing: Deep Face Generation and Editing with Disentangled Geometry and Appearance Control".ACM TRANSACTIONS ON GRAPHICS 40.4(2021):15.