20 June 2018

【论文原文】Deblur-Gan

Posted by GuoWY in Admin General
DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks

We present DeblurGAN, an end-to-end learned method for motion deblurring. The learning is based on a condi- tional GAN and the content loss . DeblurGAN achieves state-of-the art performance both in the structural similarity measure and visual appearance. The quality of the deblur- ring model is also evaluated in a novel way on a real-world problem – object detection on (de-)blurred images. The method is 5 times faster than the closest competitor – Deep- Deblur [25]. We also introduce a novel method for gen- erating synthetic motion blurred images from sharp ones, allowing realistic dataset augmentation..

More
21 June 2018

【好文推荐】令人拍案叫绝的Wasserstein GAN

Posted by GuoWY in Admin General
令人拍案叫绝的Wasserstein GAN

要知道自从2014年Ian Goodfellow提出以来,GAN就存在着训练困难、生成器和判别器的loss无法指示训练进程、生成样本缺乏多样性等问题。从那时起,很多论文都在尝试解决,但是效果不尽人意,比如最有名的一个改进DCGAN依靠的是对判别器和生成器的架构进行实验枚举,最终找到一组比较好的网络架构设置,但是实际上是治标不治本,没有彻底解决问题。而今天的主角Wasserstein GAN(下面简称WGAN)成功地做到了以下爆炸性的几点: 彻底解决GAN训练不稳定的问题,不再需要小心平衡生成器和判别器的训练程度 基本解决了collapse mode的问题,确保了生成样本的多样性 训练过程中终于有一个像交叉熵、准确率这样的数值来指示训练的进程,这个数值越小代表GAN训练得越好,代表生成器产生的图像质量越高(如题图所示) 以上一切好处不需要精心设计的网络架构,最简单的多层全连接网络就可以做到。

More
30 June 2018

【论文原文】Generative Image Inpainting with Contextual Attention

Posted by GuoWY in Admin General
Generative Image Inpainting with Contextual Attention

Recent deep learning based approaches have shown promising results for the challenging task of inpainting large missing regions in an image. These methods can generate visually plausible image structures and textures, but often create distorted structures or blurry textures in- consistent with surrounding areas. This is mainly due to ineffectiveness of convolutional neural networks in explic- itly borrowing or copying information from distant spa- tial locations. On the other hand, traditional texture and patch synthesis approaches are particularly suitable when it needs to borrow textures from the surrounding regions. Motivated by these observations, we propose a new deep generative model-based approach which can not only syn- thesize novel image structures but also explicitly utilize surrounding image features as references during network training to make better predictions.

More
03 July 2018

【论文原文】NVIDIA Image-Inpainting

Posted by GuoWY in Admin General
Image Inpainting for Irregular Holes Using Partial Convolutions

Existing deep learning based image inpainting methods use a standard convolutional network over the corrupted image, using con- volutional filter responses conditioned on both valid pixels as well as the substitute values in the masked holes (typically the mean value). This often leads to artifacts such as color discrepancy and blurriness. Post- processing is usually used to reduce such artifacts, but are expensive and may fail. We propose the use of partial convolutions, where the convolu- tion is masked and renormalized to be conditioned on only valid pixels. We further include a mechanism to automatically generate an updated mask for the next layer as part of the forward pass. Our model out- performs other methods for irregular masks. We show qualitative and quantitative comparisons with other methods to validate our approach.

More
29 July 2018

【论文原文】Glow

Posted by GuoWY in Admin General
Glow: Generative Flow with Invertible 1×1 Convolutions

Flow-based generative models (Dinh et al., 2014) are conceptually attractive due to tractability of the exact log-likelihood, tractability of exact latent-variable inference, and parallelizability of both training and synthesis. In this paper we propose Glow, a simple type of generative flow using an invertible 1 × 1 convolution. Using our method we demonstrate a significant improvement in log-likelihood on standard benchmarks. Perhaps most strikingly, we demonstrate that a generative model optimized towards the plain log-likelihood objective is capable of efficient realistic- looking synthesis and manipulation of large images.

More
07 Aug 2018

【论文原文】Beyond Face Rotation

Posted by GuoWY in Admin General
Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis

This paper proposes a Two-Pathway Generative Adversarial Network (TP-GAN) for photorealistic frontal view synthesis by simultaneously perceiving global structures and local details. Four land- mark located patch networks are proposed to attend to local textures in addition to the commonly used global encoder- decoder network. Except for the novel architecture, we make this ill-posed problem well constrained by introducing a combination of adversarial loss, symmetry loss and iden- tity preserving loss. The combined loss function leverages both frontal face distribution and pre-trained discriminative deep face models to guide an identity preserving inference of frontal views from profiles. Different from previous deep learning methods that mainly rely on intermediate features for recognition, our method directly leverages the synthe- sized identity preserving image for downstream tasks like face recognition and attribution estimation. Experimental results demonstrate that our method not only presents com- pelling perceptual results but also outperforms state-of-the- art results on large pose face recognition.

More