SADRNet: Self-Aligned Dual Face Regression Networks for Robust 3D Dense Face Alignment and Reconstruction

被引:33
|
作者
Ruan, Zeyu [1 ]
Zou, Changqing [2 ]
Wu, Longhai [3 ]
Wu, Gangshan [1 ]
Wang, Limin [1 ]
机构
[1] Nanjing Univ, State Key Lab Novel Software Technol, Nanjing 210023, Peoples R China
[2] Sun Yat Sen Univ, Sch Data & Comp Sci, Guangzhou 510006, Peoples R China
[3] Samsung Elect China R&D Ctr, Nanjing 210012, Peoples R China
基金
中国国家自然科学基金;
关键词
Faces; Three-dimensional displays; Face recognition; Shape; Image reconstruction; Geometry; Solid modeling; Three-dimensional deep face reconstruction; dense face alignment; occlusion-aware attention; MODEL; SHAPE;
D O I
10.1109/TIP.2021.3087397
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Three-dimensional face dense alignment and reconstruction in the wild is a challenging problem as partial facial information is commonly missing in occluded and large pose face images. Large head pose variations also increase the solution space and make the modeling more difficult. Our key idea is to model occlusion and pose to decompose this challenging task into several relatively more manageable subtasks. To this end, we propose an end-to-end framework, termed as Self-aligned Dual face Regression Network (SADRNet), which predicts a pose-dependent face, a pose-independent face. They are combined by an occlusion-aware self-alignment to generate the final 3D face. Extensive experiments on two popular benchmarks, AFLW2000-3D and Florence, demonstrate that the proposed method achieves significant superior performance over existing state-of-the-art methods.
引用
收藏
页码:5793 / 5806
页数:14
相关论文
共 50 条
  • [1] Joint Face Alignment and 3D Face Reconstruction with Application to Face Recognition
    Liu, Feng
    Zhao, Qijun
    Liu, Xiaoming
    Zeng, Dan
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2020, 42 (03) : 664 - 678
  • [2] Dual Neural Networks Coupling Data Regression With Explicit Priors for Monocular 3D Face Reconstruction
    Fan, Xin
    Cheng, Shichao
    Huyan, Kang
    Hou, Minjun
    Liu, Risheng
    Luo, Zhongxuan
    IEEE TRANSACTIONS ON MULTIMEDIA, 2021, 23 : 1252 - 1263
  • [3] Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network
    Feng, Yao
    Wu, Fan
    Shao, Xiaohu
    Wang, Yanfeng
    Zhou, Xi
    COMPUTER VISION - ECCV 2018, PT XIV, 2018, 11218 : 557 - 574
  • [4] Joint Face Alignment and 3D Face Reconstruction
    Liu, Feng
    Zeng, Dan
    Zhao, Qijun
    Liu, Xiaoming
    COMPUTER VISION - ECCV 2016, PT V, 2016, 9909 : 545 - 560
  • [5] 3D3M: 3D Modulated Morphable Model for Monocular Face Reconstruction
    Li, Yong
    Hao, Qiang
    Hu, Jianguo
    Pan, Xinmiao
    Li, Zechao
    Cui, Zhen
    IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 : 6642 - 6652
  • [6] 3D Face Reconstruction Based on a Single Image: A Review
    Diao, Haojie
    Jiang, Xingguo
    Fan, Yang
    Li, Ming
    Wu, Hongcheng
    IEEE ACCESS, 2024, 12 : 59450 - 59473
  • [7] 3D Face Reconstruction From A Single Image Assisted by 2D Face Images in the Wild
    Tu, Xiaoguang
    Zhao, Jian
    Xie, Mei
    Jiang, Zihang
    Balamurugan, Akshaya
    Luo, Yao
    Zhao, Yang
    He, Lingxiao
    Ma, Zheng
    Feng, Jiashi
    IEEE TRANSACTIONS ON MULTIMEDIA, 2021, 23 (23) : 1160 - 1172
  • [8] Joint Face Alignment and 3D Face Reconstruction with Efficient Convolution Neural Networks
    Li, Keqiang
    Wu, Huaiyu
    Shang, Xiuqin
    Shen, Zhen
    Xiong, Gang
    Dong, Xisong
    Hu, Bin
    Wang, Fei-Yue
    2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2021, : 6973 - 6979
  • [9] Dual Attention MobDenseNet(DAMDNet) for Robust 3D Face Alignment
    Jiang, Lei
    Wu, Xiao-Jun
    Kittler, Josef
    2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW), 2019, : 504 - 513
  • [10] Dense and Sparse 3D Deformation Signatures for 3D Dynamic Face Recognition
    Shabayek, Abd El Rahman
    Aouada, Djamila
    IEEE ACCESS, 2021, 9 : 38687 - 38705