Filters








599 Hits in 5.0 sec

Domain Bridge for Unpaired Image-to-Image Translation and Unsupervised Domain Adaptation [article]

Fabio Pizzati, Raoul de Charette, Michela Zaccaria, Pietro Cerri
2020 arXiv   pre-print
We make use of a network for clear to rain translation trained with the domain bridge to extend our work to Unsupervised Domain Adaptation (UDA).  ...  Finally, a novel approach for self-supervised learning is presented, and used to further align the domains.  ...  Thus, our innovations are spread between image-to-image translation (Sec. 3.1) and Unsupervised Domain Adaptation for semantic segmentation (Sec. 3.2).  ... 
arXiv:1910.10563v3 fatcat:mn2mtfq6m5chjmhuknfv763t6a

Domain Bridge for Unpaired Image-to-Image Translation and Unsupervised Domain Adaptation

Fabio Pizzati, Raoul de Charette, Michela Zaccaria, Pietro Cerri
2020 2020 IEEE Winter Conference on Applications of Computer Vision (WACV)  
We make use of a network for clear to rain translation trained with the domain bridge to extend our work to Unsupervised Domain Adaptation (UDA).  ...  Finally, a novel approach for self-supervised learning is presented, and used to further align the domains.  ...  Thus, our innovations are spread between image-to-image translation (Sec. 3.1) and Unsupervised Domain Adaptation for semantic segmentation (Sec. 3.2).  ... 
doi:10.1109/wacv45572.2020.9093540 dblp:conf/wacv/PizzatiCZC20 fatcat:rdroqfzeyzasdktqulb3vfmyhu

Unsupervised Domain Adaptation for Mammogram Image Classification: A Promising Tool for Model Generalization [article]

Yu Zhang, Gongbo Liang, Nathan Jacobs, Xiaoqin Wang
2020 arXiv   pre-print
In this work, we propose an unsupervised domain adaptation (UDA) method using Cycle-GAN to improve the generalization ability of the model without using any additional manual annotations.  ...  Generalization is one of the key challenges in the clinical validation and application of deep learning models to medical images.  ...  Step 1) train Cycle-GAN by using unpaired, unlabeled UKY and DDSM datasets; Step 2) translate UKY to DDSM; 3) train deep learning models by using UKY and synthesized DDSM.  ... 
arXiv:2003.01111v1 fatcat:fe76ifbgcbfvhoo4zw7qjx572y

Style-transfer GANs for bridging the domain gap in synthetic pose estimator training [article]

Pavel Rojtberg, Thomas Pöllabauer, Arjan Kuijper
2020 arXiv   pre-print
We propose to adopt general-purpose GAN models for pixel-level image translation, allowing to formulate the domain gap itself as a learning problem.  ...  The obtained models are then used either during training or inference to bridge the domain gap.  ...  Direct image domain translation In this section we introduce a CYCLEGAN [30] based training pipeline for unsupervised domain adaptation.  ... 
arXiv:2004.13681v2 fatcat:4mfmp22l35fwtbhuhpudvm5bsu

Unsupervised Image Super-Resolution with an Indirect Supervised Path

Shuaijun Chen, Zhen Han, Enyan Dai, Xu Jia, Ziluan Liu, Xing Liu, Xueyi Zou, Chunjing Xu, Jianzhuang Liu, Qi Tian
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
However, unsupervised image translation models need to be modified to adapt to unsupervised low-level vision task which poses higher requirement on the accuracy of translation.  ...  It takes the synthetic LR images as a bridge and creates an indirect supervised path.  ...  Taking it as a bridge, we are able to generate LR images ÎL real with similar noise and blur in real LR space through an unsupervised image translation model.  ... 
doi:10.1109/cvprw50498.2020.00242 dblp:conf/cvpr/ChenHDJLLZXL020 fatcat:lkq743ojb5bzdklhlpirp2sgla

Generative Damage Learning for Concrete Aging Detection using Auto-flight Images [article]

Takato Yasuno, Akira Ishii, Junichiro Fujii, Masazumi Amakata, Yuta Takahashi
2020 arXiv   pre-print
This paper proposes an anomaly detection method using unpaired image-to-image translation mapping from damaged images to reverse aging fakes that approximates healthy conditions.  ...  In order to monitor the state of large-scale infrastructures, image acquisition by autonomous flight drones is efficient for stable angle and high-quality images.  ...  Acknowledgments We would like to thank Shinich Kuramoto and Takuji Fukumoto for providing practical deep learning MATLAB resources.  ... 
arXiv:2006.15257v2 fatcat:4l3mfa4kcjer5juzztd2rhxie4

Unsupervised Image Super-Resolution with an Indirect Supervised Path [article]

Zhen Han, Enyan Dai, Xu Jia, Xiaoying Ren, Shuaijun Chen, Chunjing Xu, Jianzhuang Liu, Qi Tian
2019 arXiv   pre-print
It takes the synthetic LR images as a bridge and creates an indirect supervised path from real LR images to HR images.  ...  There are several attempts that directly apply unsupervised image translation models to address such a problem.  ...  Unsupervised image translation. Image super-resolution can be considered as a special image translation task, i.e., translating images from LR domain to HR domain.  ... 
arXiv:1910.02593v2 fatcat:fzdh5sjk2nbkpc7fdgcpkvrpoq

Learning Domain Adaptive Features with Unlabeled Domain Bridges [article]

Yichen Li, Xingchao Peng
2019 arXiv   pre-print
Conventional cross-domain image-to-image translation or unsupervised domain adaptation methods assume that the source domain and target domain are closely related.  ...  Firstly, we introduce the framework of Cycle-consistency Flow Generative Adversarial Networks (CFGAN) that utilizes domain bridges to perform image-to-image translation between two distantly distributed  ...  For unsupervised domain adaptation, we devise Prototypical Adversarial Domain Adaptation framework, which utilizes bridge domain to facilitate knowledge transfer from the source domain to a distant target  ... 
arXiv:1912.05004v1 fatcat:44b5tbq7wnhqnkrxsbbdtryjee

CrossMoDA 2021 challenge: Benchmark of Cross-Modality Domain Adaptation techniques for Vestibular Schwannoma and Cochlea Segmentation [article]

Reuben Dorent, Aaron Kujawa, Marina Ivory, Spyridon Bakas, Nicola Rieke, Samuel Joutard, Ben Glocker, Jorge Cardoso, Marc Modat, Kayhan Batmanghelich, Arseniy Belkov, Maria Baldeon Calisto (+28 others)
2022 arXiv   pre-print
All top-performing methods made use of an image-to-image translation approach to transform the source-domain images into pseudo-target-domain images.  ...  Domain Adaptation (DA) has recently raised strong interests in the medical imaging community.  ...  Acknowledgements We would like to thank all the other team members that helped during the challenge: Sewon Kim, Yohan Jun, Taejoon Eo, Dosik Hwang (Samoyed); Fei Yu, Jie Zhao, Bin Dong (PKU BIALAB); Can  ... 
arXiv:2201.02831v2 fatcat:qdd3rj62czdmxjwinmif7mkay4

Bridging the Gap between Events and Frames through Unsupervised Domain Adaptation [article]

Nico Messikommer, Daniel Gehrig, Mathias Gehrig, Davide Scaramuzza
2021 arXiv   pre-print
Our task transfer method consistently outperforms methods applicable in the Unsupervised Domain Adaptation setting for object detection by 0.26 mAP (increase by 93%) and classification by 2.7% accuracy  ...  This feature split enables to efficiently match the latent space for events and images, which is crucial for a successful task transfer.  ...  This unsupervised domain adaption is possible by leveraging labeled grayscale images and unlabeled events.  ... 
arXiv:2109.02618v1 fatcat:upjywg4szvfkfbrcippf74bfai

Neural Hair Rendering [article]

Menglei Chai, Jian Ren, Sergey Tulyakov
2020 arXiv   pre-print
Unlike existing supervised translation methods that require model-level similarity to preserve consistent structure representation for both real images and fake renderings, our method adopts an unsupervised  ...  We demonstrate the superiority of our method by testing it on a large number of portraits and comparing it with alternative baselines and state-of-the-art unsupervised image translation methods.  ...  In the context of image-to-image translation, one of the major challenges is how to bridge both the source and target domains for proper translation.  ... 
arXiv:2004.13297v2 fatcat:frk3mg3vsjfgtbaajn5uby5qty

Extremely Weak Supervised Image-to-Image Translation for Semantic Segmentation [article]

Samarth Shukla, Luc Van Gool, Radu Timofte
2019 arXiv   pre-print
In this paper, we aim to bridge the gap between supervised and unsupervised I2I translation, with application to semantic image segmentation.  ...  The current I2I translation approaches require training images from the two domains that are either all paired (supervised) or all unpaired (unsupervised).  ...  This work was partly supported by ETH General Fund and by Nvidia through a GPU grant.  ... 
arXiv:1909.08542v1 fatcat:o34fdgz4fnbqbajflrh5jnf424

StereoGAN: Bridging Synthetic-to-Real Domain Gap by Joint Optimization of Domain Translation and Stereo Matching

Rui Liu, Chengxi Yang, Wenxiu Sun, Xiaogang Wang, Hongsheng Li
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Although unsupervised image-to-image translation networks represented by CycleGAN show great potential in dealing with domain gap, it is non-trivial to generalize this method to stereo matching due to  ...  In this paper, we propose an end-to-end training framework with domain translation and stereo matching networks to tackle this challenge.  ...  , CUHK14207814, CUHK14213616, CUHK14207319, CUHK14208619, and in part by Research Impact Fund R5001-18.  ... 
doi:10.1109/cvpr42600.2020.01277 dblp:conf/cvpr/LiuYSWL20 fatcat:7r6mbp5y3fa2dp42gzyblssmpe

Deep CG2Real: Synthetic-to-Real Translation via Image Disentanglement [article]

Sai Bi, Kalyan Sunkavalli, Federico Perazzi, Eli Shechtman, Vladimir Kim, Ravi Ramamoorthi
2020 arXiv   pre-print
Training an unpaired synthetic-to-real translation network in image space is severely under-constrained and produces visible artifacts.  ...  Furthermore, networks trained on our generated "real" images predict more accurate depth and normals than domain adaptation approaches, suggesting that improving the visual realism of the images can be  ...  Acknowledgements This work was supported in part by ONR grant N000141712687, Adobe Research, a Samsung GRO grant, and the UC San Diego Center for Visual Computing.  ... 
arXiv:2003.12649v1 fatcat:6mg6ojbqcjdjfhvhxemzd7xoaa

USIS: Unsupervised Semantic Image Synthesis [article]

George Eskandar, Mohamed Abdelsamad, Karim Armanious, Bin Yang
2021 arXiv   pre-print
In this initial work, we propose a new Unsupervised paradigm for Semantic Image Synthesis (USIS) as a first step towards closing the performance gap between paired and unpaired settings.  ...  On the other hand, generic unpaired image-to-image translation frameworks underperform in comparison, because they color-code semantic layouts and feed them to traditional convolutional networks, which  ...  the efficient expansion and transformation of existing AI modules of autonomous vehicles to new domains."  ... 
arXiv:2109.14715v1 fatcat:3dgcw4ei5nc47o4e3qo47toc4a
« Previous Showing results 1 — 15 out of 599 results