AugGAN: Cross Domain Adaptation with GAN-Based Data Augmentation
By Sheng-Wei Huang, Che-Tsung Lin, Shu-Ping Chen, Yen-Yi Wu, Po-Hao Hsu, Shang-Hong Lai. 2018
256 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Generative Models
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mingsheng Long
- Jianmin Wang
- Zhangjie Cao
- Luc Van Gool
- Ming-Hsuan Yang
- Yi-Hsuan Tsai
- Jian Liang
- Michael I. Jordan
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Ming-Hsuan Yang
- Mingsheng Long
- Jianmin Wang
- Dacheng Tao
- Jiashi Feng
- Yi Yang
- Zhangjie Cao
In Generative Models
- Ming-Hsuan Yang
- Hung-Yu Tseng
- Changhee Han
- Maneesh Kumar Singh
- Changick Kim
- Michael I. Jordan
- D. Bull
- Nantheera Anantrasirichai
Citations in the corpus
- Progressive Domain Adaptation for Object Detection (2019, methodology): “For semantic segmentation, CyCADA [12] and AugGAN [14] both design a Cy-cleGAN [36]-like network to transform images from the source domain to the target one.”
- Learning More with Less: Conditional PGGAN-based Data Augmentation for Brain Metastases Detection Using Highly-Rough Annotation on MR Images (2019, background): “GANs can typically generate more realistic images than other common deep generative models, such as variational autoencoders [19] suffering from the injected noise and imperfect reconstruction due to a single objective function [22]; thus, as a DA technique, most researchers chose GANs for facilitat”
- Deep semantic segmentation of natural and medical images: a review (2019, background): “The red pixels show the ground truth and the predicted foregrounds in the left and right columns respectively.”
- Self-Ensembling With GAN-Based Data Augmentation for Domain Adaptation in Semantic Segmentation (2019, methodology): “Most previous studies for GAN-based Image-to-Image translation methods [53, 49, 27, 25, 21, 19, 31] rely on various forms of cycle-consistency.”
- A Comprehensive Survey of Image Augmentation Techniques for Deep Learning (2022, methodology): “AugGAN [52] 2018 Aim to semantically preserve object when changing its style.”
- Artificial intelligence in the creative industries: a review (2020, methodology): “g sets, with the most popular network currently being CycleGAN [22]. The original CycleGAN mapped one input to only one output, causing ineciencies when dataset diversity is required. The authors in [307] improved CycleGAN with a structure-aware network to augment training data for vehicle detection”
- Universal Domain Adaptation (2019, methodology): “Methods based on generative models synthesize labeled target samples as data augmentation and match domains in both pixel and feature levels [19, 36, 16, 20, 26, 17, 41].”