537 Hits in 7.1 sec

Exploiting Style and Attention in Real-World Super-Resolution [article]

Xin Ma, Yi Li, Huaibo Huang, Mandi Luo, Ran He
2020 arXiv   pre-print
Real-world image super-resolution (SR) is a challenging image translation problem.  ...  Our pipeline consists of a style Variational Autoencoder (styleVAE) and a SR network incorporated with attention mechanism.  ...  The SSRVAE unifies a style-based Variational Autoencoder (styleVAE) and a SR network.  ... 
arXiv:1912.10227v2 fatcat:e5c5dbbmtjdfjdtdabsw4rrybm

SinIR: Efficient General Image Manipulation with Single Image Reconstruction [article]

Jihyeong Yoo, Qifeng Chen
2021 arXiv   pre-print
We propose SinIR, an efficient reconstruction-based framework trained on a single natural image for general image manipulation, including super-resolution, editing, harmonization, paint-to-image, photo-realistic  ...  style transfer, and artistic style transfer.  ...  Perceptual losses for real-time style transfer and super-resolution. In ECCV, 2016. Karras, T., Aila, T., Laine, S., and Lehtinen, J.  ... 
arXiv:2106.07140v1 fatcat:qgkzy7c6vfhndei5r2ige77iru

Generative Adversarial Networks for Image Super-Resolution: A Survey [article]

Chunwei Tian, Xuanyu Zhang, Jerry Chun-Wen Lin, Wangmeng Zuo, Yanning Zhang
2022 arXiv   pre-print
Then, we analyze motivations, implementations and differences of GANs based optimization methods and discriminative learning for image super-resolution in terms of supervised, semi-supervised and unsupervised  ...  Single image super-resolution (SISR) has played an important role in the field of image processing.  ...  GANs on small samples for image applications 1) GANs on small samples for image style transfer: Makeup has important applications in the real world [62] .  ... 
arXiv:2204.13620v1 fatcat:hlwdqith65cxrbqrnbphjz6u4u

A survey on generative adversarial networks for imbalance problems in computer vision tasks

Vignesh Sampath, Iñaki Maurtua, Juan José Aguilar Martín, Aitor Gutierrez
2021 Journal of Big Data  
The real-world challenges and implementations of synthetic image generation based on GANs are extensively covered in this survey.  ...  In this paper, we examine the most recent developments of GANs based techniques for addressing imbalance problems in image data.  ...  Acknowledgements The authors would like to thank the anonymous reviewers for their valuable comments and suggestions on the paper.  ... 
doi:10.1186/s40537-021-00414-0 pmid:33552840 pmcid:PMC7845583 fatcat:g3p6hbjuj5c5vbe23ms4g6ed6q

Semantic Image Synthesis Manipulation for Stability Problem using Generative Adversarial Networks: A Survey

کریم مجدی, Ghada Khoriba, Hala Abbas
2021 النشرة المعلوماتیة فی الحاسبات والمعلومات  
Furthermore, opening a way to use a consistent and unified loss function for different tasks, datasets, and various generated images will be considerable assistance to tackle the challenges of training  ...  SRGAN [28] for super-resolution images and it is the first research which has been effective in achieving single image super-resolution.  ...  The result of this is that the discriminator will autoencode real samples better.  ... 
doi:10.21608/fcihib.2021.52566.1006 fatcat:dpabfqd3zbdkjjpqyp3xjljzai

Deep Generative Adversarial Networks for Image-to-Image Translation: A Review

Aziz Alotaibi
2020 Symmetry  
Image-to-image translation with generative adversarial networks (GANs) has been intensively studied and applied to various tasks, such as multimodal image-to-image translation, super-resolution translation  ...  This article provides a comprehensive overview of image-to-image translation based on GAN algorithms and its variants.  ...  Super-Resolution Super-resolution (SR) refers to the process of translating a low-resolution source image to a high-resolution target image.  ... 
doi:10.3390/sym12101705 fatcat:rqlwjjhrvbc6fhc4mxjjvkwk6i

Pose-Guided High-Resolution Appearance Transfer via Progressive Training [article]

Ji Liu, Heshan Liu, Mang-Tik Chiu, Yu-Wing Tai, Chi-Keung Tang
2020 arXiv   pre-print
We propose a novel pose-guided appearance transfer network for transferring a given reference appearance to a target pose in unprecedented image resolution (1024 * 1024), given respectively an image of  ...  our network utilizes dense local descriptors including local perceptual loss and local discriminators to refine details, which is trained progressively in a coarse-to-fine manner to produce the high-resolution  ...  The datasets we collected contains 20 hip-hop dancing videos from World of Dance competition, all of which have large pose variations.  ... 
arXiv:2008.11898v1 fatcat:r2egp7s3ejfbpes6cr7bakoofe

Deep Generative Models in Engineering Design: A Review [article]

Lyle Regenwetter, Amin Heyrani Nobari, Faez Ahmed
2022 arXiv   pre-print
Recently, DGMs such as feedforward Neural Networks (NNs), Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and certain Deep Reinforcement Learning (DRL) frameworks have shown promising  ...  Anticipating continued growth, we conduct a review of recent advances to benefit researchers interested in DGMs for design.  ...  Variational Autoencoders Introduced in 2013, Variational Autoencoders found significant success in many machine learning applications.  ... 
arXiv:2110.10863v4 fatcat:zc4mo4nwzjdlne5jbvaetyugxy

Disentangling Content and Style via Unsupervised Geometry Distillation [article]

Wayne Wu, Kaidi Cao, Cheng Li, Chen Qian, Chen Change Loy
2019 arXiv   pre-print
We evaluate our approach on four image datasets, on which we demonstrate the superior disentanglement and visual analogy quality both in synthesized and real-world data.  ...  We are able to generate photo-realistic images with 256*256 resolution that are clearly disentangled in content and style.  ...  CONCLUSION We propose a novel model based on Autoencoder framework to disentangle object's representation by content and style.  ... 
arXiv:1905.04538v1 fatcat:ssgwmthssngytohvsgxfarlseq

Auto-Embedding Generative Adversarial Networks for High Resolution Image Synthesis [article]

Yong Guo, Qi Chen, Jian Chen, Qingyao Wu, Qinfeng Shi, Mingkui Tan
2019 arXiv   pre-print
In our network, we use an autoencoder to learn the intrinsic high-level structure of real images and design a novel denoiser network to provide photo-realistic details for the generated images.  ...  However, most of the existing GAN-based methods can only produce low-resolution images of limited quality.  ...  The methods described in [14] , [27] , which are based on autoencoders, combine a Variational Autoencoder (VAE) with GAN using variational inference to solve the intractability of the marginal likelihood  ... 
arXiv:1903.11250v2 fatcat:vrb3gcfpubhx3b6zaod62fyrxa

Improved ArtGAN for Conditional Synthesis of Natural Image and Artwork [article]

Wei Ren Tan, Chee Seng Chan, Hernan Aguirre, Kiyoshi Tanaka
2018 arXiv   pre-print
Inspired by recent works, an autoencoder is incorporated into the categorical discriminator for additional complementary information.  ...  Qualitatively, we demonstrate that ArtGAN is able to generate plausible-looking images on Oxford-102 and CUB-200, as well as able to draw realistic artworks based on style, artist, and genre.  ...  ACKNOWLEDGMENT We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan X GPU used for this research.  ... 
arXiv:1708.09533v2 fatcat:glppdbaxwvfvdn5nd4wpkyrk4u

Perceptual Image Super-Resolution with Progressive Adversarial Network [article]

Lone Wong, Deli Zhao, Shaohua Wan, Bo Zhang
2020 arXiv   pre-print
To address this issue, we propose Progressive Adversarial Network (PAN) that is capable of coping with this difficulty for domain-specific image super-resolution.  ...  Single Image Super-Resolution (SISR) aims to improve resolution of small-size low-quality image from a single one.  ...  Super-resolution of real-world low resolution faces in arbitrary poses with GANs (Super-FAN) [6] and Multi-Task Fig. 1 : Illustration of the curse of dimensionality.  ... 
arXiv:2003.03756v4 fatcat:dg32vyec5ndhrhmpin7kp4uhwi

Generative Adversarial Networks (GANs): An Overview of Theoretical Model, Evaluation Metrics, and Recent Developments [article]

Pegah Salehi, Abdolah Chalechale, Maryam Taghizadeh
2020 arXiv   pre-print
The generator learns to generate plausible data, and the discriminator learns to distinguish fake data created by the generator from real data samples.  ...  | Image super-resolution Image super-resolution (SR) has been widely used in satellite, medical, and military images, and more.  ...  | Developments Based on Autoencoders Autoencoder neural networks are a type of deep neural networks used for feature extraction and reconstruction operations.  ... 
arXiv:2005.13178v1 fatcat:u57mmd76njef7nclxgeyzycziy

Super-Resolution of Sentinel-2 Images Using a Spectral Attention Mechanism

Maialen Zabalza, Angela Bernardini
2022 Remote Sensing  
In this work, we implement a state-of-the-art residual learning-based model called Super-Resolution Residual Network (SRResNet), which we train using PlanetScope-Sentinel pairs of images.  ...  However, the spatial resolution of these satellites is insufficient for many tasks.  ...  Moreover, since most of the existing SISR methods have been implemented using synthetic data, their super-resolution performance is drastically altered when using real-world images [7] .  ... 
doi:10.3390/rs14122890 fatcat:46thrcfbkrcmhjofz3oqmjz2fu

Good Artists Copy, Great Artists Steal: Model Extraction Attacks Against Image Translation Generative Adversarial Networks [article]

Sebastian Szyller, Vasisht Duddu, Tommi Gröndahl, N. Asokan
2021 arXiv   pre-print
(super resolution).  ...  In this paper, we show the first model extraction attack against real-world generative adversarial network (GAN) image translation models.  ...  As previously explained, training paired translation models is not realistic in many real world applications due to the lack of styled images.  ... 
arXiv:2104.12623v1 fatcat:sp5hwosgxjexjhbq2dcydetwba
« Previous Showing results 1 — 15 out of 537 results