|
|
论文:2023,Vol:41,Issue(2):370-378 |
|
|
引用本文： |
|
|
刘洋, 樊养余, 马浩悦, 吕国云, 刘诗雅. 一种基于可见性权重的人脸纹理融合方法[J]. 西北工业大学学报 |
|
|
LIU Yang, FAN Yangyu, MA Haoyue, LYU Guoyun, LIU Shiya. A method of face texture fusion based on visibility weight[J]. Journal of Northwestern Polytechnical University |
|
|
|
|
|
|
|
| 一种基于可见性权重的人脸纹理融合方法 |
|
| 刘洋1, 樊养余1, 马浩悦1, 吕国云1, 刘诗雅2 |
|
1. 西北工业大学 电子信息学院, 陕西 西安 710072;
2. 虚拟现实内容制作中心, 北京 101318 |
| 摘要: |
| 由二维图像到三维人脸模型的重建过程中,不同角度下的人脸在进行纹理补全时,存在纹理像素模糊和亮度差异的问题,提出一种基于可见性权重的方法,对于不同视角下的人脸纹理图进行融合。同时,针对耳部区域结构复杂,传统纹理融合算法不适用的情况,利用高斯模型的肤色概率方法对耳部纹理缺失区域进行补全,并设计纹理融合带平滑像素,最终得到完整、逼真度较高的人脸纹理模型。仿真实验结果表明,所提人脸纹理融合与补全方法实现了多角度下头面部区域纹理的完美融合。与目前通用的方法相比较,在同样的渲染条件下,所提方法生成的人脸纹理模型更完整、纹理显示效果更好。 |
| 关键词:
纹理融合
纹理补全
人脸重建
纹理映射
|
|
| A method of face texture fusion based on visibility weight |
|
| LIU Yang1, FAN Yangyu1, MA Haoyue1, LYU Guoyun1, LIU Shiya2 |
|
1. School of Electronics and Information, Northwestern Polytechnical University, Xi'an 710072, China;
2. Content Production Center of Virtual Reality, Beijing 101318, China |
| Abstract: |
| In the reconstruction process from 2D images to 3D face models, texture completion still suffers from pixel blurring and color inconsistency when face images are under different perspectives. In this paper, we propose a method based on visibility weights for face texture fusion. Meanwhile, for the complex geometric structure of the ear region where the traditional texture mapping algorithm is inapplicable, a skin color probability method with Gaussian model is used for pixel completion, and jointly optimized with the texture fusion band. Finally, we generate a complete and high-fidelity face texture model. The simulation experiment shows that the novel face texture fusion and completion method generates the perfect texture under multiple viewpoints. Our face texture model outperforms state-of-the-art techniques under the same rendering conditions. |
| Key words:
face texture completion
texture fusion
face reconstruction
face texture mapping
|
|
| 收稿日期: 2022-05-23
修回日期:
|
| DOI: 10.1051/jnwpu/20234120370 |
| 基金项目: 陕西省重点研发计划(2020ZDLGY04-09)资助 |
| 通讯作者: 吕国云(1975-),西北工业大学副教授,主要从事图像处理、虚拟现实研究。e-mail:lvguoyun101@nwpu.edu.cn
Email：lvguoyun101@nwpu.edu.cn |
| 作者简介: 刘洋(1992-),西北工业大学博士研究生,主要从事计算机图形图像处理、虚拟人研究。
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] JAMES B, ANASTASIOS R, STEFANOS Z, et al. A 3D morphable model learnt from 10 000 faces[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR), 2016
[2] 吴斌, 孙显, 王宏琦, 等. 一种三维建筑物模型自动纹理映射方法[J]. 遥感信息, 2017, 32(2):5 WU Bin, SUN Xian, WANG Hongqi, et al. A method of automatic texture mapping for 3D reconstruction model[J]. Remote Sensing Information, 2017, 32(2):5 (in Chinese)
[3] 曾成强. 纹理映射技术算法综述[J]. 甘肃科技纵横, 2014, 43(11):40-44 ZENG Chengqiang. A survey of texture mapping algorithms[J]. Gansu Science and Technology, 2014, 43(11):40-44 (in Chinese)
[4] BIER E A, SLOAN K R. Two-part texture mappings[J]. IEEE Computer Graphics and Application, 1986, 6(9):40-53
[5] DEEPAK P, PHILIPP K, JEFF D, et al. Context encoder:feature learning by inpainting[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016
[6] LIU Guilin, FITSUM A R, KEVIN J S, et al. Image inpainting for irregular holes using partial convolutions[C]//Proceedings of the European Conference on Computer Vision, 2018
[7] DENG Jiankang, CHENG Shiyang, XUE Niannan, et al. UV-GAN:adversarial facial UV map completion for pose-invariant face recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018
[8] 刘洋, 樊养余, 郭哲, 等. 单幅人脸图像的全景纹理图生成方法[J]. 中国图象图形学报, 2022, 27(2):602-613 LIU Yang, FAN Yangyu, GUO Zhe, et al. Solo face image-based panoramic texture map generation[J]. Journal of Image and Graphics, 2022, 27(2):602-613 (in Chinese)
[9] YAMAGUCHI S, SAITO S, NAGANO K, et al. High-fidelity facial reflectance and geometry inference from an unconstrained image[J]. ACM Transactions on Graphics, 2018, 37(4):162
[10] LATTAS A, STYLIANOS M, GECER B, et al. AvatarMe:realistically renderable 3D facial reconstruction[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2020
[11] MOLLER T, TRUMBORE B. Fast, minimum storage ray/triangle intersection[J]. Journal of Graphics Tools, 1997, 2(1):21-28
[12] GECER B, DENG J, ZAFEIRIOS S. OSTeC:one-shot texture completion[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2021
[13] ZENG X, WU Z, PENG X, et al. Joint 3D facial shape reconstruction and texture completion from a single image[J]. Computational Visual Media, 2022, 8(2):239-256
[14] RAMON E, TRIGINER G, ESCUR J, et al. H3D-Net:Few-shot high-fidelity 3D head reconstruction[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021 |
|
|
|
|
|
|
|
