基于生成对抗网络的多角度人脸重构
Multi-Poses Face Reconstruction Based on Generative Adversarial Networks
DOI: 10.12677/CSA.2020.103046, PDF,    科研立项经费支持
作者: 赵志舜*, 程良伦, 黄国恒：广东工业大学，计算机学院，广东 广州
关键词: 人脸重构；人脸转正；生成对抗网络；Face Reconstruction； Face Frontalization； Generative Adversarial Networks
摘要: 由于在不同角度下人脸的差异性很大，多角度的人脸重构仍然是一项具有挑战性的研究课题。现存的方法只能建立两个不同人脸角度之间的映射。若要生成多角度人脸图像，只能通过多个生成器进行多次训练来实现任意两个角度之间的映射，这极大地增加了训练的时间成本。本文提出了一种基于生成对抗网络的多角度人脸重构算法。通过引入对抗损失函数和循环一致性损失函数，本文中的算法仅通过一次训练就能建立任意两个人脸角度之间的映射，这极大地降低了训练的时间成本以及减少了训练难度。在公开数据集Multi-PIE上进行对比实验，结果表明本算法不仅能进行多角度人脸转换，而且在人脸转正之后的人脸识别准确率明显高于CPF、DR-GAN、Light-CNN等算法。
Abstract: Because of the large discrepancy between human faces at different angles, multi-poses face recon-struction is still a challenging research topic. Existing methods can only establish a mapping of two different face poses. To generate multi-poses face images, multiple generators have to be trained multiple times to achieve the mapping between any two poses, which greatly increases the time cost of training. In this paper, a face reconstruction algorithm based on generative adversarial networks is proposed. By introducing a combination of adversarial loss and cycle consistency loss, the algorithm in this paper can establish a mapping between any two face poses through only one training, which greatly reduces the time cost and the difficulty of training. Comparative experiments on the public dataset Multi-PIE show that this algorithm can not only perform multi-poses face generation, the accuracy of face recognition after face frontalization is significantly higher than CPF, DR-GAN, Light-CNN and other algorithms.
文章引用：赵志舜, 程良伦, 黄国恒. 基于生成对抗网络的多角度人脸重构[J]. 计算机科学与应用, 2020, 10(3): 445-455. https://doi.org/10.12677/CSA.2020.103046

参考文献

[1] Guo, Y., Zhang, L., Hu, Y., He, X., et al. (2016) Ms-celeb-1m: A Dataset and Benchmark for Large-Scale Face Recognition. European Conference on Computer Vision, Amsterdam, 8-16 October 2016, 87-102. [Google Scholar] [CrossRef] 
[2] Liu, Z., Luo, P., Wang, X. and Tang, X. (2015) Deep Learning Face Attributes in the Wild. Proceedings of the IEEE International Conference on Computer Vision, Santiago, 11-18 December 2015, 3730-3738. [Google Scholar] [CrossRef] 
[3] Oh, B.S., Toh, K.A., Teoh, A.B.J., et al. (2018) An Analytic Gabor Feed forward Network for Single-Sample and Pose-Invariant Face Recognition. IEEE Transactions on Image Pro-cessing, 27, 2791-2805. [Google Scholar] [CrossRef] 
[4] Zhang, Z., Chen, X., Wang, B., et al. (2018) Face Frontalization Using an Appearance-Flow-Based Convolutional Neural Network. IEEE Transactions on Image Processing, 28, 2187-2199. [Google Scholar] [CrossRef] 
[5] Tan, X. and Triggs, B. (2010) Enhanced Local Tex-ture Feature Sets for Face Recognition under Difficult Lighting Conditions. IEEE Transactions on Image Processing, 19, 1635-1650. [Google Scholar] [CrossRef] 
[6] Banerji, S., Sinha, A. and Liu, C. (2013) New Image Descriptors Based on Color, Texture, Shape, and Wavelets for Object and Scene Image Classification. Neuro-computing, 117, 173-185. [Google Scholar] [CrossRef] 
[7] Chen, J., Shan, S., He, C., et al. (2009) WLD: A Robust Local Image Descriptor. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32, 1705-1720. [Google Scholar] [CrossRef] 
[8] Hu, J., Lu, J. and Tan, Y.P. (2015) Deep Transfer Metric Learning. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, 8-10 June 2015, 325-333. [Google Scholar] [CrossRef] 
[9] Tran, L., Yin, X. and Liu, X. (2018) Rep-resentation Learning by Rotating Your Faces. IEEE Transactions on Pattern Analysis and Machine Intelligence, 41, 3007-3021. [Google Scholar] [CrossRef] 
[10] Yim, J., Jung, H., Yoo, B., et al. (2015) Rotating Your Face Using Multi-Task Deep Neural Network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, 8-10 June 2015, 676-684.
[11] Goodfellow, I., Pouget-Abadie, J., Mirza, M., et al. (2014) Generative Adversarial Nets. Proceedings of the 27th International Conference on Neural Information Processing Systems, Volume 2, December 2014, 2672-2680.
[12] Tran, L., Yin, X. and Liu, X. (2017) Disentangled Representation Learning Gan for Pose-Invariant Face Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Hawaii, 21-26 July 2017, 1415-1424. [Google Scholar] [CrossRef] 
[13] Huang, R., Zhang, S., Li, T., et al. (2017) Beyond Face Rotation: Global and Local Perception Gan for Photorealistic and Identity Preserving Frontal View Synthesis. Proceedings of the IEEE International Conference on Computer Vision, Venice, 22-29 October 2017, 2439-2448. [Google Scholar] [CrossRef] 
[14] Chen, Y., Lai, Y.K. and Liu, Y.J. (2018) Cartoongan: Generative Adversarial Networks for Photo Cartoonization. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, 18-22 June 2018, 9465-9474. [Google Scholar] [CrossRef] 
[15] Xiao, T., Hong, J. and Ma, J. (2018) Elegant: Exchanging Latent Encodings with Gan for Transferring Multiple Face Attributes. Proceedings of the European Conference on Computer Vision, Munich, 8-14 September 2018, 168-184. [Google Scholar] [CrossRef] 
[16] Wang, X., Yu, K., Wu, S., et al. (2018) Esrgan: Enhanced Super-Resolution Generative Adversarial Networks. Proceedings of the European Conference on Computer Vision, Munich, 8-14 September 2018, 63-79. [Google Scholar] [CrossRef] 
[17] Radford, A., Metz, L. and Chintala, S. (2015) Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks.
https://arxiv.org/abs/1511.06434
[18] Choi, Y., Choi, M., Kim, M., et al. (2018) Stargan: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation. Proceedings of the IEEE Conference on Com-puter Vision and Pattern Recognition, Salt Lake City, 18-22 June 2018, 8789-8797. [Google Scholar] [CrossRef] 
[19] Ghodrati, A., Jia, X., Pedersoli, M., et al. (2015) Towards Automatic Image Editing: Learning to See Another You.
https://arxiv.org/abs/1511.08446
[20] Zhu, X., Lei, Z., Yan, J., et al. (2015) High-Fidelity Pose and Expression Normalization for Face Recognition in the Wild. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, 8-10 June 2015, 787-796.
[21] Zhu, Z., Luo, P., Wang, X., et al. (2013) Deep Learning Identi-ty-Preserving Face Space. Proceedings of the IEEE International Conference on Computer Vision, Sydney, 3-6 De-cember 2013, 113-120. [Google Scholar] [CrossRef] 
[22] Zhu, Z., Luo, P., Wang, X., et al. (2014) Multi-View Perceptron: A Deep Model for Learning Face Identity and View Representations. Advances in Neural Information Processing Systems, Vol. 1, 217-225.
[23] Wu, X., He, R., Sun, Z., et al. (2018) A Light CNN for Deep Face Representation with Noisy Labels. IEEE Transactions on Information Forensics and Security, 13, 2884-2896. [Google Scholar] [CrossRef]