DeCGAN: a diversity-enhanced CycleGAN for unpaired image-to-image translation [PDF]
Scientific ReportsUnpaired image-to-image translation, demonstrated by frameworks like CycleGAN, has shown promise in critical applications, including medical imaging, but is hindered by challenges such as mode collapse and structural degradation.
Matthew Cobbinah +10 more
doaj +4 more sources
Haar Wavelet-Based Representation Learning for Unpaired Image-to-Image Translation
IEEE AccessIn recent years, there have been numerous attempts to achieve unpaired image-to-image translation. Many algorithms have especially incorporated the contrastive learning framework into unpaired image-to-image translation. This paper presents an innovative
Soobin Park +3 more
doaj +4 more sources
Exploring the Possibility of Measuring Vertebrae Bone Structure Metrics Using MDCT Images: An Unpaired Image-to-Image Translation Method [PDF]
Bioengineering, 2023 Bone structure metrics are vital for the evaluation of vertebral bone strength. However, the gold standard for measuring bone structure metrics, micro-Computed Tomography (micro-CT), cannot be used in vivo, which hinders the early diagnosis of fragility ...
Dan Jin, Han Zheng, Huishu Yuan
doaj +2 more sources
Clinical Potential of Artificial Bone Scintigraphy from Early-Phase Bone Scintigraphy Using Unpaired Image-to-Image Translation in Patients with Breast Cancer: A Single-Center Prospective Study [PDF]
TomographyObjectives: The objective of this study is to investigate the clinical potential of generating artificial bone scintigraphy (aBS), defined here as a deep learning-generated bone scintigraphy image that simulates delayed-phase bone scintigraphy (dBS ...
Yong-Jin Park +4 more
doaj +2 more sources
Unpaired image to image translation for source free domain adaptation in semantic segmentation [PDF]
Scientific ReportsSource-free domain adaptation (SFDA) assumes that source data are inaccessible during domain adaptation. Current SFDA methods commonly utilize source-trained models to generate pseudolabels for unlabelled target data.
Juan An +7 more
doaj +2 more sources
Multi‐head mutual‐attention CycleGAN for unpaired image‐to‐image translation
IET Image Processing, 2020 The image‐to‐image translation, i.e. from source image domain to target image domain, has made significant progress in recent years. The most popular method for unpaired image‐to‐image translation is CycleGAN.
Wei Ji, Jing Guo, Yun Li
doaj +2 more sources
A domain translation network with contrastive constraint for unpaired motion image deblurring
IET Image Processing, 2023 Most motion deblurring methods require a large amount of paired training data, which is nearly unreachable in practice. To overcome the limitation, a domain translation network with contrastive constraint for unpaired motion image deblurring is proposed.
Bingxin Zhao, Weihong Li
doaj +2 more sources
Depth-aware unpaired image-to-image translation for autonomous driving test scenario generation using a dual-branch GAN [PDF]
Frontiers in NeuroroboticsReliable visual perception is essential for autonomous driving test scenario generation, yet adverse weather and lighting variations pose significant challenges to simulation robustness and generalization.
Donghao Shi +20 more
doaj +2 more sources
Enhancing generative networks for chest anomaly localization through automatic registration-based unpaired-to-pseudo-paired training data translation [PDF]
Scientific ReportsGenerative adversarial network–based image translation (GAN-IT) has emerged as a promising approach for chest X-ray (CXR) anomaly localization (AL-CXR) without pixel-level annotations, especially when training relies on unpaired data.
Seong Je Oh +6 more
doaj +2 more sources
Multi-attention bidirectional contrastive learning method for unpaired image-to-image translation [PDF]
PLoS ONEBenchen Yang +4 more
doaj +2 more sources