Semantics-Preserving Sketch Embedding for Face Generation

被引:2
作者
Yang, Binxin [1 ]
Chen, Xuejin [1 ]
Wang, Chaoqun [1 ]
Zhang, Chi [1 ]
Chen, Zihan [1 ]
Sun, Xiaoyan [1 ]
机构
[1] Univ Sci & Technol China, Elect Engn & Informat Sci, Hefei 230026, Peoples R China
关键词
Semantics; Faces; Aerospace electronics; Codes; Image synthesis; Task analysis; Generators; Sketch-based generation; face generation; image-to-image translation; semantics-preserving;
D O I
10.1109/TMM.2023.3239182
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
With recent advances in image-to-image translation tasks, remarkable progress has been witnessed in generating face images from sketches. However, existing methods frequently fail to generate images with details that are semantically and geometrically consistent with the input sketch, especially when various decoration strokes are drawn. To address this issue, we introduce a novel W-W+ encoder architecture to take advantage of the high expressive power of W+ space and semantic controllability of W space. We introduce an explicit intermediate representation for sketch semantic embedding. With a semantic feature matching loss for effective semantic supervision, our sketch embedding precisely conveys the semantics in the input sketches to the synthesized images. Moreover, a novel sketch semantic interpretation approach is designed to automatically extract semantics from vectorized sketches. We conduct extensive experiments on both synthesized sketches and hand-drawn sketches, and the results demonstrate the superiority of our method over existing approaches on both semantics-preserving and generalization ability.
引用
收藏
页码:8657 / 8671
页数:15
相关论文
共 46 条
  • [1] StyleFlow: Attribute-conditioned Exploration of StyleGAN-Generated Images using Conditional Continuous Normalizing Flows
    Abdal, Rameen
    Zhu, Peihao
    Mitra, Niloy J.
    Wonka, Peter
    [J]. ACM TRANSACTIONS ON GRAPHICS, 2021, 40 (03):
  • [2] Image2StyleGAN: How to Embed Images Into the StyleGAN Latent Space?
    Abdal, Rameen
    Qin, Yipeng
    Wonka, Peter
    [J]. 2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, : 4431 - 4440
  • [3] Alaluf Y., 2022, P IEEE CVF C COMP VI, P18511
  • [4] ReStyle: A Residual-Based StyleGAN Encoder via Iterative Refinement
    Alaluf, Yuval
    Patashnik, Or
    Cohen-Or, Daniel
    [J]. 2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), 2021, : 6691 - 6700
  • [5] Barrow H. G., 1977, IJCAI, V2, P659
  • [6] DeepFaceEditing: Deep Face Generation and Editing with Disentangled Geometry and Appearance Control
    Chen, Shu-Yu
    Liu, Feng-Lin
    Lai, Yu-Kun
    Rosin, Paul L.
    Li, Chunpeng
    Fu, Hongbo
    Gao, Lin
    [J]. ACM TRANSACTIONS ON GRAPHICS, 2021, 40 (04):
  • [7] DeepFaceDrawing: Deep Generation of Face Images from Sketches
    Chen, Shu-Yu
    Su, Wanchao
    Gao, Lin
    Xia, Shihong
    Fu, Hongbo
    [J]. ACM TRANSACTIONS ON GRAPHICS, 2020, 39 (04):
  • [8] SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis
    Chen, Wengling
    Hays, James
    [J]. 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 9416 - 9425
  • [9] CHUNG J, 2014, NIPS 2014 WORKSH DEE, DOI DOI 10.48550/ARXIV.1412.3555
  • [10] Du J, 2018, Arxiv, DOI [arXiv:1710.10370, DOI 10.48550/ARXIV.1710.10370]