A Self-Supervised Bootstrap Method for Single-Image 3D Face Reconstruction

被引:3
|
作者
Xing, Yifan [1 ]
Tewari, Rahul [1 ]
Mendonc, Paulo R. S. [1 ]
机构
[1] Amazon Web Serv, Seattle, WA 98109 USA
来源
2019 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV) | 2019年
关键词
DATABASE;
D O I
10.1109/WACV.2019.00113
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
State-of-the-art methods for 3D reconstruction of faces from a single image require 2D-3D pairs of ground-truth data for supervision. Such data is costly to acquire, and most datasets available in the literature are restricted to pairs for which the input 2D images depict faces in a near fronto-parallel pose. Therefore, many data-driven methods for single-image 3D facial reconstruction perform poorly on profile and near-profile faces. We propose a method to improve the performance of single-image 3D facial reconstruction networks by utilizing the network to synthesize its own training data for fine-tuning, comprising: (i) single-image 3D reconstruction of faces in near-frontal images without ground-truth 3D shape; (ii) application of a rigid-body transformation to the reconstructed face model; (iii) rendering of the face model from new viewpoints; and (iv) use of the rendered image and corresponding 3D reconstruction as additional data for supervised fine-tuning. The new 2D-3D pairs thus produced have the same high-quality observed for near fronto-parallel reconstructions, thereby nudging the network towards more uniform performance as a function of the viewing angle of input faces. Application of the proposed technique to the fine-tuning of a state-of-the-art single-image 3D-reconstruction network for faces demonstrates the usefulness of the method, with particularly significant gains for profile or near-profile views.
引用
收藏
页码:1014 / 1023
页数:10
相关论文
共 50 条
  • [31] Curriculum Self-Supervised Learning for 3D CT Cardiac Image Segmentation
    Taher, Mohammad Reza Hosseinzadeh
    Ikuta, Masaki
    Soni, Ravi
    MACHINE LEARNING FOR HEALTH, ML4H, VOL 225, 2023, 225 : 145 - 156
  • [32] Self-supervised 3D face reconstruction based on multi-scale feature fusion and dual attention mechanism
    Zhou D.-K.
    Zhang C.
    Yang X.
    Jilin Daxue Xuebao (Gongxueban)/Journal of Jilin University (Engineering and Technology Edition), 2022, 52 (10): : 2428 - 2437
  • [33] LOW-FREQUENCY GUIDED SELF-SUPERVISED LEARNING FOR HIGH-FIDELITY 3D FACE RECONSTRUCTION IN THE WILD
    Wang, Pengrui
    Lin, Chunze
    Xu, Bo
    Che, Wujun
    Wang, Quan
    2020 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO (ICME), 2020,
  • [34] Deep Self-Supervised Hyperspectral Image Reconstruction
    Liu, Zhe
    Han, Xian-Hua
    ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS, 2022, 18 (03)
  • [35] Self-supervised reflectance-guided 3d shape reconstruction from single-view images
    Fang, Binbin
    Xiao, Nanfeng
    APPLIED INTELLIGENCE, 2023, 53 (06) : 6966 - 6977
  • [36] Self-supervised reflectance-guided 3d shape reconstruction from single-view images
    Binbin Fang
    Nanfeng Xiao
    Applied Intelligence, 2023, 53 : 6966 - 6977
  • [37] Single-Image 3D Human Pose and Shape Estimation Enhanced by Clothed 3D Human Reconstruction
    Liu, Leyuan
    Gao, Yunqi
    Sun, Jianchi
    Chen, Jingying
    ARTIFICIAL INTELLIGENCE AND ROBOTICS, ISAIR 2023, 2024, 1998 : 33 - 44
  • [38] A self-supervised method of single-image depth estimation by feeding forward information using max-pooling layers
    Shi, Jinlong
    Sun, Yunhan
    Bai, Suqin
    Sun, Zhengxing
    Tian, Zhaohui
    VISUAL COMPUTER, 2021, 37 (04): : 815 - 829
  • [39] Animating Landscape: Self-Supervised Learning of Decoupled Motion and Appearance for Single-Image Video Synthesis
    Endo, Yuki
    Kanamori, Yoshihiro
    Kuriyama, Shigeru
    ACM TRANSACTIONS ON GRAPHICS, 2019, 38 (06):
  • [40] A self-supervised method of single-image depth estimation by feeding forward information using max-pooling layers
    Jinlong Shi
    Yunhan Sun
    Suqin Bai
    Zhengxing Sun
    Zhaohui Tian
    The Visual Computer, 2021, 37 : 815 - 829