DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks

Kupyn, Orest; Budzan, Volodymyr; Mykhailych, Mykola; Mishkin, Dmytro; Matas, Jiřı́

doi:10.1109/cvpr.2018.00854

Cited by 1,419 publications

(1,305 citation statements)

References 35 publications

Supporting

Mentioning

1,302

Contrasting

Unclassified

Order By: Relevance

“…Table 1 Quantitative comparison with the state-of-the-art methods on super resolving the blurry images. The evaluated methods include super resolution methods, SRResNet , EDSR , RCAN , image deblurring methods, DeepDeblur , DeblurGAN (Kupyn et al, 2018), SRN , and joint approaches, SCGAN , ED-DSRN . The methods with a sign are trained on our LR-GOPRO training set.…”

Section: Super Resolving Rainy Imagementioning

confidence: 99%

Gated Fusion Network for Degraded Image Super Resolution

Zhang

Dong

Hu³

et al. 2020

Int J Comput Vis

View full text Add to dashboard Cite

Single image super resolution aims to enhance image quality with respect to spatial content, which is a fundamental task in computer vision. In this work, we address the task of single frame super resolution with the presence of image degradation, e.g., blur, haze, or rain streaks. Due to the limitations of frame capturing and formation processes, image degradation is inevitable, and the artifacts would be exacerbated by super resolution methods. To address this problem, we propose a dual-branch convolutional neural network to extract base features and recovered features separately. The base features contain local and global information of the input image. On the other hand, the recovered features focus on the degraded regions and are used to remove the degradation. Those features are then fused through a recursive gate module to obtain sharp features for super resolution. By decomposing the feature extraction step into two task-independent streams, the dual-branch model can facilitate the training process by avoiding learning the mixed degradation all-in-one and thus enhance the final high-resolution prediction results. We evaluate the proposed method in three degradation scenarios. Experiments on these scenarios demonstrate that the proposed method performs more efficiently and favorably against the state-of-the-art approaches on benchmark datasets.

show abstract

Section: Super Resolving Rainy Imagementioning

confidence: 99%

Gated Fusion Network for Degraded Image Super Resolution

Zhang

Dong

Hu³

et al. 2020

Int J Comput Vis

View full text Add to dashboard Cite

show abstract

“…For the deblurring and optical flow modules, we take advantage of existing neural network architectures which have performed well in the past for the respective supervised learning tasks [9], [17], [21], [30], [32]. In particular, we adopt the single image deblurring network from Tao et al [32] and the dense optical flow estimation network PWC-Net from Sun et al [30].…”

Section: A Deblurring and Optical Flowmentioning

confidence: 99%

“…We compare the single image deblurring results of our network quantitatively and qualitatively with a state-of-the-art optimization-based method [39], supervised methods [17], [21], [32] as well as the domain specific self-supervised method from [19]. We train all networks with their recommended hyperparameter settings on our synthetic dataset.…”

Section: Baselines and Experimental Settingsmentioning

confidence: 99%

“…As many computer vision algorithms such as semantic segmentation, object detection, or visual odometry rely on visual input, blurry images challenge the performance of these algorithms. It is well known that many algorithms (e.g., depth prediction, feature detection, motion estimation, or object recognition) suffer from motion blur [17], [25], [26], [33]. The motion deblurring problem has thus received considerable attention in the past [7], [17], [21], [28], [32].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Self-Supervised Linear Motion Deblurring

Liu

Janai

Pollefeys

et al. 2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Motion blurry images challenge many computer vision algorithms, e.g., feature detection, motion estimation, or object recognition. Deep convolutional neural networks are stateof-the-art for image deblurring. However, obtaining training data with corresponding sharp and blurry image pairs can be difficult. In this paper, we present a differentiable reblur model for selfsupervised motion deblurring, which enables the network to learn from real-world blurry image sequences without relying on sharp images for supervision. Our key insight is that motion cues obtained from consecutive images yield sufficient information to inform the deblurring task. We therefore formulate deblurring as an inverse rendering problem, taking into account the physical image formation process: we first predict two deblurred images from which we estimate the corresponding optical flow. Using these predictions, we re-render the blurred images and minimize the difference with respect to the original blurry inputs. We use both synthetic and real dataset for experimental evaluations. Our experiments demonstrate that self-supervised single image deblurring is really feasible and leads to visually compelling results. Both the code and datasets are available at https://github.com/ethliup/SelfDeblur.

show abstract

“…Different applications have been proposed in recent years, including skin cancer detection 9 in medical areas, crack detection for many construction tasks, 25,26 and photo postprocessing for many smartphone cameras. 27,28 A typical DCNN prototype can be found in Figure 1. In DCNN, the features are learned by stacking multiple layers of convolution and pooling operations.…”

Section: Deep Learning In Computer Visionmentioning

confidence: 99%

Deep learning for pyrolysis reactor monitoring: From thermal imaging toward smart monitoring system

Zhu

Benton

et al. 2018

AIChE Journal

View full text Add to dashboard Cite

Monitoring the operation of a pyrolysis reactor is always challenging due to the extremely high‐operating temperature (over 800°C) in the fired furnace. To improve current monitoring capability, a monitoring framework is proposed that builds upon thermal photography to provide a detailed view inside the fired furnace. Based on the infrared images generated from the temperature data provided by cameras, a deep learning approach is introduced to automatically identify tube regions from the raw images. The pixel‐wise tube segmentation network is named Res50‐UNet, which combines the popular ResNet‐50 and U‐Net architectures. By this approach, the precise temperature and shape on pyrolysis tubes are monitored. The control limits are eventually drawn by the adaptive k‐nearest neighbor method to raise alarms for faults. Through testing over real plant data, the framework assists process operators by providing in‐depth operating information of the reactor and fault diagnosis. © 2018 American Institute of Chemical Engineers AIChE J, 65: 582–591, 2019

show abstract

DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks

Cited by 1,419 publications

References 35 publications

Gated Fusion Network for Degraded Image Super Resolution

Gated Fusion Network for Degraded Image Super Resolution

Self-Supervised Linear Motion Deblurring

Deep learning for pyrolysis reactor monitoring: From thermal imaging toward smart monitoring system

Contact Info

Product

Resources

About