Generative Adversarial Networks with Inverse Transformation Unit.
作者： Shuo Ding , Zhifeng Kong
DOI:
关键词:
摘要: In this paper we introduce a new structure to Generative Adversarial Networks by adding an inverse transformation unit behind the generator. We present two theorems claim convergence of model, and conjectures nonideal situations when is not bijection. A general survey on models with different transformations was done MNIST dataset Fashion-MNIST dataset, which shows does necessarily need be Also, certain that blurs image, our model successfully learned sharpen images recover blurred images, additionally verified measurement sharpness.
arxiv.org
harvard.edu参考文章(13)
Y. Lecun, L. Bottou, Y. Bengio, P. Haffner, Gradient-based learning applied to document recognition Proceedings of the IEEE. ,vol. 86, pp. 2278- 2324 ,(1998) , 10.1109/5.726791
Soumith Chintala, Alec Radford, Luke Metz, Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks arXiv: Learning. ,(2015)
Noah Snavely, Kavita Bala, Paul Upchurch, From A to Z: Supervised Transfer of Style and Content Using Deep Neural Network Generators arXiv: Computer Vision and Pattern Recognition. ,(2016)
Shumeet Baluja, Learning Typographic Style arXiv: Computer Vision and Pattern Recognition. ,(2016)
Botond Cseke, Sebastian Nowozin, Ryota Tomioka, f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization arXiv: Machine Learning. ,(2016)
Junbo Zhao, Michael Mathieu, Yann LeCun, None, Energy-based Generative Adversarial Network arXiv: Learning. ,(2016)
Yong Yu, Lantao Yu, Weinan Zhang, Jun Wang, SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient arXiv: Learning. ,(2016)
Yaniv Taigman, Lior Wolf, Adam Polyak, Unsupervised Cross-Domain Image Generation arXiv: Computer Vision and Pattern Recognition. ,(2016)
Guo-Jun Qi, Loss-Sensitive Generative Adversarial Networks on Lipschitz Densities arXiv: Computer Vision and Pattern Recognition. ,(2017)
Phillip Isola, Alexei A. Efros, Jun-Yan Zhu, Taesung Park, Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks arXiv: Computer Vision and Pattern Recognition. ,(2017)