A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf
.
Filters
Application of Deep Learning in Blind Motion Deblurring: Current Status and Future Prospects
[article]
2024
arXiv
pre-print
Following this, we categorize and summarize existing blind motion deblurring methods based on different backbone networks, including convolutional neural networks, generative adversarial networks, recurrent ...
Motion deblurring is one of the fundamental problems of computer vision and has received continuous attention. ...
[90] proposed an unsupervised image deblurring method using a multi-adversarial optimization CycleGAN to address artifacts in high-resolution image generation. ...
arXiv:2401.05055v1
fatcat:s2syuu42avaw3otjebrarp3nme
Uncertainty-Aware Variate Decomposition for Self-supervised Blind Image Deblurring
2023
Proceedings of the 31st ACM International Conference on Multimedia
To overcome this limitation, unsupervised deblurring methods have been proposed by using natural priors or generative adversarial networks. ...
Extensive comparisons demonstrate that the proposed framework outperforms state-of-theart unsupervised methods on both dynamic scene, human-aware centric motion, real-world and out-of-focus deblurring ...
Human-aware Centric Motion Deblurring. Compared with dynamic scene deblurring, human-aware centric motion deblurring further considers differences between foreground and background. ...
doi:10.1145/3581783.3612535
fatcat:hdf2ac7wznaxjejrqgelslny2y
When Autonomous Systems Meet Accuracy and Transferability through AI: A Survey
[article]
2020
arXiv
pre-print
Finally, we discuss several challenges and future topics for using adversarial learning, RL and meta-learning in autonomous systems. ...
in autonomous systems, including image style transfer, image superresolution, image deblurring/dehazing/rain removal, semantic segmentation, depth estimation, pedestrian detection and person re-identification ...
[266] proposed a multi-city navigation network with LSTM structure. ...
arXiv:2003.12948v3
fatcat:qtmjs74p2vh6thdotbhgebdvoi
Unsupervised Domain-Specific Deblurring using Scale-Specific Attention
[article]
2021
arXiv
pre-print
To address the above issues, we propose unsupervised domain-specific deblurring using a scale-adaptive attention module (SAAM). ...
Our network does not require supervised pairs for training, and the deblurring mechanism is primarily guided by adversarial loss, thus making our network suitable for a distribution of blur functions. ...
These instances from a single image using adversarial loss.
priors are used to estimate the underlying latent image and
camera motion using alternating minimization techniques. ...
arXiv:2112.06175v1
fatcat:rzz2t7dkonbaxp3ylx4li3alpm
When Autonomous Systems Meet Accuracy and Transferability through AI: A Survey
2020
Patterns
Finally, we discuss several challenges and future topics for the use of adversarial learning, RL, and meta-learning in autonomous systems. ...
vision tasks in autonomous systems, including image style transfer, image super-resolution, image deblurring/dehazing/rain removal, semantic segmentation, depth estimation, pedestrian detection, and person ...
Generative Adversarial Networks for Semantic Segmentation and Multi-Task (A) CycleGAN for semantic segmentation Copyright (2017) IEEE. ...
doi:10.1016/j.patter.2020.100050
pmid:33205114
pmcid:PMC7660378
fatcat:vs7wm2yrwjamjbaml36663wvze
FCL-GAN: A Lightweight and Real-Time Baseline for Unsupervised Blind Image Deblurring
[article]
2022
arXiv
pre-print
Blind image deblurring (BID) remains a challenging and significant task. ...
In this paper, we propose a lightweight and real-time unsupervised BID baseline, termed Frequency-domain Contrastive Loss Constrained Lightweight CycleGAN (shortly, FCL-GAN), with attractive properties ...
[21] first try to estimate the global camera motion from small-scale paired data in a semi-supervised manner and use the obtained global camera motion to perform single image deblurring and change detection ...
arXiv:2204.07820v2
fatcat:2xvksrgwobetxfpqjmy3mgngj4
2020 Index IEEE Transactions on Computational Imaging Vol. 6
2020
IEEE Transactions on Computational Imaging
., +, TCI 2020 1127-1138 Fast Multi-Focus Ultrasound Image Recovery Using Generative Adversarial Networks. ...
., +, TCI 2020 1127-1138 Fast Multi-Focus Ultrasound Image Recovery Using Generative Adversarial Networks. ...
doi:10.1109/tci.2021.3054596
fatcat:puij7ztll5ai7alxrmqzsupcny
Semantic information supplementary pyramid network for dynamic scene deblurring
2020
IEEE Access
In addition, SIS-net uses the intermediate layer path to extract image features in a single time to obtain a multi-scale effect. ...
We choose Generative Adversarial Network (GAN) as its fundamental model. ...
[1] builds a generative adversarial model to remove motion blur. ...
doi:10.1109/access.2020.3028157
fatcat:ktwsy42adrcplkeg6rridacqji
2021 Index IEEE Transactions on Image Processing Vol. 30
2021
IEEE Transactions on Image Processing
., +, TIP 2021 4905-4918 Structure-Aware Motion Deblurring Using Multi-Adversarial Optimized CycleGAN. ...
Ali, U., +, TIP 2021 7215-7227 Structure-Aware Motion Deblurring Using Multi-Adversarial Optimized CycleGAN. ...
doi:10.1109/tip.2022.3142569
fatcat:z26yhwuecbgrnb2czhwjlf73qu
Neural Global Shutter: Learn to Restore Video from a Rolling Shutter Camera with Global Reset Feature
[article]
2022
arXiv
pre-print
This feature enables us to turn the rectification problem into a deblur-like one, getting rid of inaccurate and costly explicit motion estimation. ...
The widely used rolling-shutter (RS) image sensors, however, suffer from geometric distortion when the camera and object undergo motion during capture. ...
, GS motion deblur (MT-deblur) [11] , and GS out-of-focus deblur (OF-deblur) [13] . ...
arXiv:2204.00974v2
fatcat:6ounddd5inftdpuygloveu6q74
LODE: Deep Local Deblurring and A New Benchmark
[article]
2021
arXiv
pre-print
Then, we propose a novel framework, termed BLur-Aware DEblurring network (BladeNet), which contains three components: the Local Blur Synthesis module generates locally blurred training pairs, the Local ...
While recent deep deblurring algorithms have achieved remarkable progress, most existing methods focus on the global deblurring problem, where the image blur mostly arises from severe camera shake. ...
[23] propose a multi-scale CNN with adversarial loss in a coarse-to-fine manner. Gao et al. [6] propose a principled parameter selective sharing scheme. Zhang et al. ...
arXiv:2109.09149v1
fatcat:ckwicbqeavglldloemr7smsbxq
Learning Disentangled Feature Representation for Hybrid-distorted Image Restoration
[article]
2020
arXiv
pre-print
We also propose a feature aggregation module (FAM) with channel-wise attention to adaptively filter out the distortion representations and aggregate useful content information from different channels for ...
Zhang, H., Patel, V.M.: Density-aware single image de-raining using a multi-stream dense network. ...
Overview of Whole Framework To make full use of feature information of different levels, we use multi-phases structure to deal with hybrid distortion as Fig. 5 . ...
arXiv:2007.11430v1
fatcat:tb27oellmzegjaqavddprwob4m
A deep learning framework for quality assessment and restoration in video endoscopy
2020
Medical Image Analysis
Generative adversarial networks with carefully chosen regularization and training strategies for discriminator-generator networks are finally used to restore corrupted frames. ...
The restoration models for blind deblurring, saturation correction and inpainting demonstrate significant improvements over previous methods. ...
Computation used the Oxford Biomedical Research Computing (BMRC) facility, a joint development between the Wellcome Centre for Human Genetics and the Big Data Institute supported by Health Data Research ...
doi:10.1016/j.media.2020.101900
pmid:33246229
fatcat:4i5pqs27tfd3za3ikzq5bfw6aq
Talking Face Generation by Conditional Recurrent Adversarial Network
[article]
2019
arXiv
pre-print
In addition, we deploy a multi-task adversarial training scheme in the context of video generation to improve both photo-realism and the accuracy for lip synchronization. ...
We propose a novel conditional video generation network where the audio input is treated as a condition for the recurrent adversarial network such that temporal dependency is incorporated to realize smooth ...
The proposed method is trained using an adversarial training scheme to avoid blur result. Multi-task discriminator is applied here to both improve the lip sync accuracy as well as photo-realism. ...
arXiv:1804.04786v3
fatcat:uh42fov5wvcs5nopm4tiqnqdgq
A Survey of Deep Learning-Based Image Restoration Methods for Enhancing Situational Awareness at Disaster Sites: The Cases of Rain, Snow and Haze
2022
Sensors
This survey article is concerned with the emergence of vision augmentation AI tools for enhancing the situational awareness of first responders (FRs) in rescue operations. ...
[147] wanted their network to have the ability to handle uneven and dense haze concentration and introduced the Dark Channel Attention optimized CycleGAN (DCA-CycleGAN). ...
The multi-loss function is composed of a combination of adversary loss, L 1 loss, the structural similarity index (SSIM) loss and a new peak-signalto-noise ratio (PSNR) loss. ...
doi:10.3390/s22134707
pmid:35808203
pmcid:PMC9269588
fatcat:d34wmmvjznfcxfmh3vkj5qlk44
« Previous
Showing results 1 — 15 out of 50 results