Filters








5,591 Hits in 6.3 sec

From Motion Blur to Motion Flow: a Deep Learning Solution for Removing Heterogeneous Motion Blur [article]

Dong Gong, Jie Yang, Lingqiao Liu, Yanning Zhang, Ian Reid, Chunhua Shen, Anton van den Hengel, Qinfeng Shi
2016 arXiv   pre-print
Removing pixel-wise heterogeneous motion blur is challenging due to the ill-posed nature of the problem.  ...  Our approach directly estimates the motion flow from the blurred image through a fully-convolutional deep neural network (FCN) and recovers the unblurred image from the estimated motion flow.  ...  To handle general heterogeneous motion blur, based on the motion flow model, we propose a deep neural network based method able to directly estimate a pixel-wise motion flow map from a single blurred image  ... 
arXiv:1612.02583v1 fatcat:f4gy7sn3mngqxbs23pxrf466um

From Motion Blur to Motion Flow: A Deep Learning Solution for Removing Heterogeneous Motion Blur

Dong Gong, Jie Yang, Lingqiao Liu, Yanning Zhang, Ian Reid, Chunhua Shen, Anton Van Den Hengel, Qinfeng Shi
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Removing pixel-wise heterogeneous motion blur is challenging due to the ill-posed nature of the problem.  ...  Our approach directly estimates the motion flow from the blurred image through a fully-convolutional deep neural network (FCN) and recovers the unblurred image from the estimated motion flow.  ...  Conclusion In this paper, we proposed a flexible and efficient deep learning based method for estimating and removing the heterogeneous motion blur.  ... 
doi:10.1109/cvpr.2017.405 dblp:conf/cvpr/GongYLZRSHS17 fatcat:xlypuqjr4rbtrn2gqqly74jvze

Blur Invariant Kernel-Adaptive Network for Single Image Blind deblurring [article]

Sungkwon An, Hyungmin Roh, Myungjoo Kang
2020 arXiv   pre-print
To use the kernel efficiently, we propose a kernel-adaptive AE block that encodes features from both blurred images and blur kernels into a low dimensional space and then decodes them simultaneously to  ...  The network learns the blur pattern of the input image and trains to generate the estimation of image-specific blur kernels.  ...  In particular, to solve the motion deblurring problem, we estimated the pixel-wise motion flow, such that our model can remove pixel-wise heterogeneous motion blurs. Gong et al.  ... 
arXiv:2007.04543v3 fatcat:mhyenqhhjnd4roap6tzqkmo5za

Prior-enlightened and Motion-robust Video Deblurring [article]

Ya Zhou, Jianfeng Xu, Kazuyuki Tasaka, Zhibo Chen, Weiping Li
2020 arXiv   pre-print
Therefore, we propose a PRiOr-enlightened and MOTION-robust video deblurring model (PROMOTION) suitable for challenging blurs.  ...  Besides the classical camera shake caused global blurry, we also prove the generalization for the downstream task suffering from local blur.  ...  Another spatially variant RNN for dynamic scene deblurring is proposed in [15] , where the weights of the RNN are learned by a deep CNN.  ... 
arXiv:2003.11209v2 fatcat:7idg4c4b5beftpt5kblcsrg44a

Bringing Alive Blurred Moments [article]

Kuldeep Purohit, Anshul Shah, A. N. Rajagopalan
2019 arXiv   pre-print
We present a solution for the goal of extracting a video from a single motion blurred image to sequentially reconstruct the clear views of a scene as beheld by the camera during the time of exposure.  ...  Once trained, it is employed for guided training of a motion encoder for blurred images.  ...  The potential of our work can be extended in a variety of directions including blur-based segmentation, video deblurring, video interpolation, action recognition etc.  ... 
arXiv:1804.02913v2 fatcat:2fv7f7bsazbqjieysood5jitza

Self-Supervised Flow Estimation using Geometric Regularization with Applications to Camera Image and Grid Map Sequences [article]

Sascha Wirges, Johannes Gräter, Qiuhao Zhang, Christoph Stiller
2019 arXiv   pre-print
Furthermore, we scale the errors due to spatial flow inconsistency by a mask that we derive from the motion mask.  ...  We extend existing approaches for self-supervised optical flow estimation by adding a regularizer expressing motion consistency assuming a static environment.  ...  CONCLUSION We presented our approach to learn flow estimators based on deep convolutional networks in a self-supervised setting.  ... 
arXiv:1904.12599v1 fatcat:p5wbch7qtffgpiq2nmefwawza4

High-throughput multiparametric imaging flow cytometry: toward diffraction-limited sub-cellular detection and monitoring of sub-cellular processes

Gregor Holzner, Bogdan Mateescu, Daniel van Leeuwen, Gea Cereghetti, Reinhard Dechant, Stavros Stavrakis, Andrew deMello
2021 Cell Reports  
We present a sheathless, microfluidic imaging flow cytometer that incorporates stroboscopic illumination for blur-free fluorescence detection at ultra-high analytical throughput.  ...  Results highlight the utility of our imaging flow cytometer in directly investigating phase-separated compartments within cellular environments and screening rare events at the sub-cellular level for a  ...  The ''motion blur'' lines shown in Figure S2B define a parameter space for a given experiment.  ... 
doi:10.1016/j.celrep.2021.108824 pmid:33691119 fatcat:umsor5cfpnelzdibtssqifpa34

Virtual-freezing fluorescence imaging flow cytometry

Hideharu Mikami, Makoto Kawaguchi, Chun-Jung Huang, Hiroki Matsumura, Takeaki Sugimura, Kangrui Huang, Cheng Lei, Shunnosuke Ueno, Taichi Miura, Takuro Ito, Kazumichi Nagasawa, Takanori Maeno (+10 others)
2020 Nature Communications  
Here we present an optomechanical imaging method that overcomes the trade-off by virtually freezing the motion of flowing cells on the image sensor to effectively achieve 1000 times longer exposure time  ...  The availability of numerous information-rich fluorescence cell images allows high-dimensional statistical analysis and accurate classification with deep learning, as evidenced by our demonstration of  ...  Code availability Our image analysis codes are available at https://github.com/MortisHuang/VIFFI-imageanalysis (codes for images of cells) and https://github.com/hideharu-mikami/VIFFIflbeads (codes for  ... 
doi:10.1038/s41467-020-14929-2 pmid:32139684 fatcat:buew3pwtyrbx7e4j7i4by74udm

Virtual-freezing fluorescence imaging flow cytometry [article]

Hideharu Mikami, Makoto Kawaguchi, Chun-Jung Huang, Hiroki Matsumura, Takeaki Sugimura, Kangrui Huang, Cheng Lei, Shunnosuke Ueno, Taichi Miura, Takuro Ito, Kazumichi Nagasawa, Takanori Maeno (+10 others)
2020 bioRxiv   pre-print
Here we present an optomechanical imaging method that overcomes the trade-off by virtually "freezing" the motion of flowing cells on the image sensor to effectively achieve 1,000 times longer exposure  ...  The availability of numerous information-rich fluorescence cell images allows high-dimensional statistical analysis and accurate classification with deep learning, as evidenced by our demonstration of  ...  (e.g., deep learning) to make better decisions in biomedical research and clinical diagnosis 11, 12 .  ... 
doi:10.1101/2020.01.23.916452 fatcat:e4b4au7ugndttde7v6yfkhtybu

FMD-cGAN: Fast Motion Deblurring using Conditional Generative Adversarial Networks [article]

Jatin Kumar and Indra Deep Mastan and Shanmuganathan Raman
2021 arXiv   pre-print
Like other deep neural network architectures, GANs also suffer from large model size (parameters) and computations.  ...  In this paper, we present a Fast Motion Deblurring-Conditional Generative Adversarial Network (FMD-cGAN) that helps in blind motion deblurring of a single image.  ...  D.Gong,J.Yang,L.Liu,Y.Zhang,I.Reid,C.Shen,A.VanDenHengel,Q.Shi: From motion blur to motion flow: a deep learn- ing solution for removing heterogeneous motion blur. IEEE, 2017 5.  ... 
arXiv:2111.15438v2 fatcat:ouphc37dp5avnhxrrvgx47dbge

Human-Aware Motion Deblurring [article]

Ziyi Shen, Wenguan Wang, Xiankai Lu, Jianbing Shen, Haibin Ling, Tingfa Xu, Ling Shao
2020 arXiv   pre-print
This paper proposes a human-aware deblurring model that disentangles the motion blur between foreground (FG) humans and background (BG).  ...  It learns a soft mask that encodes FG human information and explicitly drives the FG/BG decoder-branches to focus on their specific domains.  ...  However, the diacritical mechanism relies heavily on the method of segmentation and fails to learn a robust solution for multi-motion superposition in real dynamic scenes.  ... 
arXiv:2001.06816v1 fatcat:iihosraewzhahebknotnrtzcwq

Adaptive Single Image Deblurring [article]

Maitreya Suin, Kuldeep Purohit, A. N. Rajagopalan
2022 arXiv   pre-print
Although end-to-end fully convolutional designs have recently advanced the state-of-the-art in non-uniform motion deblurring, their performance-complexity trade-off is still sub-optimal.  ...  In this work, we propose an efficient pixel adaptive and feature attentive design for handling large blur variations within and across different images.  ...  From motion blur to motion flow: a deep learning solution for removing heterogeneous motion blur. In The IEEE conference on computer vision and pattern recognition (CVPR), 2017.  ... 
arXiv:2201.00155v1 fatcat:jexcg67nqfgefi63zja27aj2xe

SegFlow: Joint Learning for Video Object Segmentation and Optical Flow [article]

Jingchun Cheng, Yi-Hsuan Tsai, Shengjin Wang, Ming-Hsuan Yang
2017 arXiv   pre-print
The unified framework is trained iteratively offline to learn a generic notion, and fine-tuned online for specific objects.  ...  This paper proposes an end-to-end trainable network, SegFlow, for simultaneously predicting pixel-wise object segmentation and optical flow in videos.  ...  For optical flow, we first use the MPI Sintel dataset [6] that contains 1041 pairs of images in synthesized scenes, with a Clean version containing images without motion blur and atmospheric effects,  ... 
arXiv:1709.06750v1 fatcat:2bc4r6te4jbwph4oksctn2ezgi

Infrared Image Deblurring Based on Generative Adversarial Networks

Yuqing Zhao, Guangyuan Fu, Hongqiao Wang, Shaolei Zhang, Min Yue
2021 International Journal of Optics  
Because the blur is not only caused by the motion of different objects but also by the relative motion and jitter of cameras, there is a change of scene depth.  ...  Different from the previous work, we combine the traditional blind deblurring method and the blind deblurring method based on the learning method, and uniform and nonuniform blurred images are considered  ...  For example, Gupta et al. [27] proposed to model camera motion as a motion density function. e blurring kernel of spatial variables can be derived directly from it.  ... 
doi:10.1155/2021/9946809 doaj:fbe4237c05254c68bcfb02b382114d56 fatcat:hu3jhmdb3bc5xkdzk6cstqpp7m

SegFlow: Joint Learning for Video Object Segmentation and Optical Flow

Jingchun Cheng, Yi-Hsuan Tsai, Shengjin Wang, Ming-Hsuan Yang
2017 2017 IEEE International Conference on Computer Vision (ICCV)  
The unified framework is trained iteratively offline to learn a generic notion, and fine-tuned online for specific objects.  ...  This paper proposes an end-to-end trainable network, SegFlow, for simultaneously predicting pixel-wise object segmentation and optical flow in videos.  ...  For optical flow, we first use the MPI Sintel dataset [6] that contains 1041 pairs of images in synthesized scenes, with a Clean version containing images without motion blur and atmospheric effects,  ... 
doi:10.1109/iccv.2017.81 dblp:conf/iccv/ChengTW017 fatcat:zvpgxvwylfby3bm747cpbqoodm
« Previous Showing results 1 — 15 out of 5,591 results