每日论文 Learning from Simulated and Unsupervised Images through Adversarial Training
2017-04-05 16:46
686 查看
苹果首发
用没有标签的真实人眼数据通过训练一个模拟器提炼网络（生成器）去提炼生成的人眼图片，增加生成人眼图片的真实性。用一个对抗损失和一个自正则化损失函数去优化训练过程。
这个优化方程优化生成器R，同时增加了约束项，将提炼的人眼和生成的人眼就行L1范式最小化。
这个优化方法优化辨别器，除了用生成的人眼数据使得值最小，还用到真实的人眼数据优化使得值最大。
为了增加真实效果，还进行了局部辨别器处理和缓存提炼图片等算法。网络也用到残差网络保留很多输入图像信息。
用没有标签的真实人眼数据通过训练一个模拟器提炼网络（生成器）去提炼生成的人眼图片，增加生成人眼图片的真实性。用一个对抗损失和一个自正则化损失函数去优化训练过程。
这个优化方程优化生成器R，同时增加了约束项，将提炼的人眼和生成的人眼就行L1范式最小化。
这个优化方法优化辨别器，除了用生成的人眼数据使得值最小，还用到真实的人眼数据优化使得值最大。
为了增加真实效果，还进行了局部辨别器处理和缓存提炼图片等算法。网络也用到残差网络保留很多输入图像信息。
内容来自用户分享和网络整理，不保证内容的准确性，如有侵权内容，可联系管理员处理 
相关文章推荐
- [Paper note] Learning from Simulated and Unsupervised Images through Adversarial Training
- 每日论文InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial
- 论文阅读笔记之《Learning Deep Correspondence through Prior and Posterior Feature Constancy》
- 论文阅读：Fast and Accurate Semiautomatic Segmentation of Individual Teeth from Dental CT Images
- 《Unsupervised Learning of Depth and Ego-Motion from Video》读书笔记
- 生成对抗网络学习笔记3----论文unsupervised representation learning with deep convolutional generative adversarial
- 论文笔记：unsupervised representation learning with deep convolutional generative adversarial networks
- [Stereo_unsupervised][cvpr16]Unsupervised learning of disparity maps from stereo images
- 读论文-Transfer from Simulation to Real World through Learning Deep Inverse Dynamics Model
- 读论文笔记：Unsupervised Joint Object Discovery and Segmentation in Internet Images
- 深度学习论文理解2:on random weights and unsupervised feature learning
- PredNet --- Deep Predictive coding networks for video prediction and unsupervised learning --- 论文笔记
- 论文阅读：Individual tooth segmentation from CT images scanned with contacts of maxillary and mandible te
- (论文分析) Machine Learning -- Learning from labeled and unlabeled data
- 论文笔记之：UNSUPERVISED REPRESENTATION LEARNING WITH DEEP CONVOLUTIONAL GENERATIVE ADVERSARIAL NETWORKS
- 论文阅读：Tooth and Alveolar Bone Segmentation from Dental Computed Tomography Images
- [导入]Produce GIF or PNG barcode images from a Ruby on Rails application using RMagick and Gbarcode
- 论文《Learning both Weights and Connections for Efficient Neural Network》阅读笔记
- Show and Tell Lessons learned from the 2015 MSCOCO Image Captioning Challenge论文及tensorflow源码解读(2)
- 论文笔记之：MatchNet: Unifying Feature and Metric Learning for Patch-Based Matching