• Corpus ID: 6281601

Generating images with recurrent adversarial networks

@article{Im2016GeneratingIW,
  title={Generating images with recurrent adversarial networks},
  author={Daniel Jiwoong Im and Chris Dongjoo Kim and Hui Jiang and Roland Memisevic},
  journal={ArXiv},
  year={2016},
  volume={abs/1602.05110}
}
Gatys et al. (2015) showed that optimizing pixels to match features in a convolutional network with respect reference image features is a way to render images of high visual quality. We show that unrolling this gradient-based optimization yields a recurrent computation that creates images by incrementally adding onto a visual "canvas". We propose a recurrent generative model inspired by this view, and show that it can be trained using adversarial training to generate very good image samples. We… 
Generating Images Part by Part with Composite Generative Adversarial Networks
TLDR
A model called composite generative adversarial network is proposed, that reveals the complex structure of images with multiple generators in which each generator generates some part of the image to create a new single image.
Deep Feature Similarity for Generative Adversarial Networks
TLDR
This work proposes a new way to train generative adversarial networks (GANs) based on pretrained deep convolutional neural network (CNN), and enforce the deep feature similarity of the generated and real images to stabilize the training and generate more natural visual images.
GENERATIVE ADVERSARIAL NETWORKS
TLDR
Stochastic clipping is introduced, a simple, gradient-based technique called stochastic clipping that precisely recover their latent vector pre-images 100% of the time and appears to recover unique encodings for unseen images.
Wasserstein Generative Recurrent Adversarial Networks for Image Generating
TLDR
A framework named the Wasserstein generative recurrent adversarial networks (WGRAN) is proposed, which merges Wassersteins distance with recurrent neural networks to iteratively generate realistic looking images and trains the model in an adversarial way.
Image Completion of Highly Noisy Images Using Deep Learning
TLDR
A novel approach that leverages the combination of capsule networks and generative adversarial networks that is used for generating images in a more realistic manner to generate images which are locally and globally consistent in nature by leveraging the low level and high-level features of an image.
LR-GAN: Layered Recursive Generative Adversarial Networks for Image Generation
TLDR
The experiments demonstrate that LR-GAN can generate more natural images with objects that are more human recognizable than DCGAN and is trained in an end-to-end manner with gradient descent methods.
Learning from Simulated and Unsupervised Images through Adversarial Training
TLDR
This work develops a method for S+U learning that uses an adversarial network similar to Generative Adversarial Networks (GANs), but with synthetic images as inputs instead of random vectors, and makes several key modifications to the standard GAN algorithm to preserve annotations, avoid artifacts, and stabilize training.
CapsGAN: Using Dynamic Routing for Generative Adversarial Networks
TLDR
CapsGAN is presented, a novel technique for generating images in the 3D domain from images with high degree of geometrical transformations using rotated MNIST and the efficacy of using capsules architecture in the GANs domain is shown.
Generative Adversarial Networks with Inverse Transformation Unit Zhifeng Kong
TLDR
A new structure to Generative Adversarial Networks is introduced by adding an inverse transformation unit behind the generator to claim the convergence of the model, and two conjectures to nonideal situations when the transformation is not bijection are presented.
How Generative Adversarial Nets and its variants Work: An Overview of GAN
TLDR
Details of GAN are looked into that firstly show how it operates and fundamental meaning of objective functions and point to GAN variants applied to vast amount of tasks.
...
...

References

SHOWING 1-10 OF 22 REFERENCES
Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
TLDR
A generative parametric model capable of producing high quality samples of natural images using a cascade of convolutional networks within a Laplacian pyramid framework to generate images in a coarse-to-fine fashion.
Conditional generative adversarial nets for convolutional face generation
TLDR
An extension of generative adversarial networks (GANs) to a conditional setting is applied, and the likelihood of real-world faces under the generative model is evaluated, and how to deterministically control face attributes is examined.
Conditional Generative Adversarial Nets
TLDR
The conditional version of generative adversarial nets is introduced, which can be constructed by simply feeding the data, y, to the generator and discriminator, and it is shown that this model can generate MNIST digits conditioned on class labels.
DRAW: A Recurrent Neural Network For Image Generation
TLDR
The Deep Recurrent Attentive Writer neural network architecture for image generation substantially improves on the state of the art for generative models on MNIST, and, when trained on the Street View House Numbers dataset, it generates images that cannot be distinguished from real data with the naked eye.
Generative Adversarial Nets
We propose a new framework for estimating generative models via an adversarial process, in which we simultaneously train two models: a generative model G that captures the data distribution, and a
Striving for Simplicity: The All Convolutional Net
TLDR
It is found that max-pooling can simply be replaced by a convolutional layer with increased stride without loss in accuracy on several image recognition benchmarks.
LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
TLDR
This work proposes to amplify human effort through a partially automated labeling scheme, leveraging deep learning with humans in the loop, and constructs a new image dataset, LSUN, which contains around one million labeled images for each of 10 scene categories and 20 object categories.
Adaptive deconvolutional networks for mid and high level feature learning
TLDR
A hierarchical model that learns image decompositions via alternating layers of convolutional sparse coding and max pooling, relying on a novel inference scheme that ensures each layer reconstructs the input, rather than just the output of the layer directly beneath, as is common with existing hierarchical approaches.
A Neural Algorithm of Artistic Style
TLDR
This work introduces an artificial system based on a Deep Neural Network that creates artistic images of high perceptual quality and offers a path forward to an algorithmic understanding of how humans create and perceive artistic imagery.
Denoising Criterion for Variational Auto-Encoding Framework
TLDR
It is shown that injecting noise both in input and in the stochastic hidden layer can be advantageous and a modified variational lower bound is proposed as an improved objective function in this setup and it is found that the proposed denoising variational autoencoder (DVAE) yields better average log-likelihood than the VAE and the importance weighted autoencoders on the MNIST and Frey Face datasets.
...
...