A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Filters
Burst Denoising with Kernel Prediction Networks
2018
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
Preserved Fulltext
In particular, we propose a convolutional neural network architecture for predicting spatially varying kernels that can both align and denoise frames, a synthetic data generation approach based on a realistic ...
We present a technique for jointly denoising bursts of images taken from a handheld camera. ...
Predicted kernels Our network predicts a stack of 2D kernels at each pixel which we visualize in Fig. 7 . ...
doi:10.1109/cvpr.2018.00265
dblp:conf/cvpr/MildenhallBCSNC18
fatcat:pkz7emcrujfpvej5ifdyzklrcm
Burst Denoising with Kernel Prediction Networks
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In particular, we propose a convolutional neural network architecture for predicting spatially varying kernels that can both align and denoise frames, a synthetic data generation approach based on a realistic ...
We present a technique for jointly denoising bursts of images taken from a handheld camera. ...
Predicted kernels Our network predicts a stack of 2D kernels at each pixel which we visualize in Fig. 7 . ...
arXiv:1712.02327v2
fatcat:wvjxs7ysbbdqbaq3idmtiqxgz4
Basis Prediction Networks for Effective Burst Denoising with Large Kernels
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
To this end, we introduce a novel basis prediction network that, given an input burst, predicts a set of global basis kernels -- shared within the image -- and the corresponding mixing coefficients -- ...
This allows us to effectively exploit comparatively larger denoising kernels, achieving both significant quality improvements (over 1dB PSNR) and faster run-times over state-of-the-art methods. ...
). input burst using a "kernel prediction network" (KPN). ...
arXiv:1912.04421v2
fatcat:3symba4cpzb6dkeyfbluhth3iq
Multi-Kernel Prediction Networks for Denoising of Burst Images
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We propose a deep neural network based approach called Multi-Kernel Prediction Networks (MKPN) for burst image denoising. ...
Recent approaches for image denoising aim to predict kernels which are convolved with a set of successively taken images (burst) to obtain a clear image. ...
Given a noisy set of input burst images, kernels are predicted for each pixel using a deep neural network. ...
arXiv:1902.05392v1
fatcat:2k3kvawaorbgpg3tyevfsdhszy
Attention Mechanism Enhanced Kernel Prediction Networks for Denoising of Burst Images
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In this paper, attention mechanism enhanced kernel prediction networks (AME-KPNs) are proposed for burst image denoising, in which, nearly cost-free attention modules are adopted to first refine the feature ...
Simulations and real-world experiments are conducted to illustrate the robustness of the proposed AME-KPNs in burst image denoising. ...
(c) (d) (e) (f)
CONCLUSION Novel attention mechanism enhanced kernel prediction networks (AME-KPNs) have been proposed for burst image denoising. ...
arXiv:1910.08313v2
fatcat:66ppnqgb6ncgdoawgauz7sl4hi
Deep Denoising of Flash and No-Flash Pairs for Photography in Low-Light Environments
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2020 pre-print arXiv:2012.05116v1 fatcat:crvmz4sqojhz5d5kqobeoonypq
We introduce a neural network-based method to denoise pairs of images taken in quick succession, with and without a flash, in low-light environments. ...
Our network outputs a gain map and a field of kernels, the latter obtained by linearly mixing elements of a per-image low-rank kernel basis. ...
Here, our network predicts kernels to be used to filter and sum all the no-flash images, which is then multiplied with our scale map. ...
arXiv:2012.05116v2
fatcat:iag6usixvrcfddflsxp5es2xwq
Digital Gimbal: End-to-end Deep Image Stabilization with Learnable Exposure Times
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2021 pre-print arXiv:2012.04515v3 fatcat:zaeh5yy6ojec5pclgozyq262yeOther Versions
We demonstrate this method's advantage over the traditional approach of deblurring a single image or denoising a fixed-exposure burst on both synthetic and real data. ...
To exploit the trade-off between motion blur at long exposures and low SNR at short exposures, we train a CNN that estimates a sharp high-SNR image by aggregating a burst of noisy short-exposure frames ...
[26] propose an adaptive kernel prediction network for jointly aligning and merging noisy frames. ...
arXiv:2012.04515v4
fatcat:vsczuc7ovvcprpn3uhyal3nhnu
Efficient Burst Raw Denoising with Variance Stabilization and Multi-frequency Denoising Network
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Denoising based on a burst of multiple frames generally outperforms single frame denoising but with the larger compututional cost. ...
Instead, we resort to a conventional and efficient alignment method and combine it with our multi-frame denoising network. ...
KPN [36] proposes kernel prediction network to jointly conduct multi-frame alignment and denoising. MPKN [34] extends single kernel prediction to multiple kernels prediction. ...
arXiv:2205.04721v1
fatcat:yslxm2bymfgcpc5brv2lhrgexe
Deep Motion Blur Removal Using Noisy/Blurry Image Pairs
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Recent progress in deep neural networks suggests that kernel free single image deblurring can be efficiently performed, but questions about deblurring performance persist. ...
Thus, we propose to restore a sharp image by fusing a pair of noisy/blurry images captured in a burst. ...
Comparison with Burst Denoising Methods Burst image denoising is an alternative strategy to the noisy/blurry image pair deblurring. ...
arXiv:1911.08541v2
fatcat:untm7ndhn5ar3bui3hi45jyh5i
Burst Image Restoration and Enhancement
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
burst low-light image enhancement, and burst denoising tasks. ...
In comparison to existing works that usually follow a late fusion scheme with single-stage upsampling, our approach performs favorably, delivering state-of-the-art performance on burst superresolution, ...
[39] generate per-pixel kernels through the kernel prediction network (KPN) to merge the input images. ...
arXiv:2110.03680v2
fatcat:6cwkgx2t4fc4jnkz33wu55mhcy
Deep Learning on Image Denoising: An overview
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
However, there are substantial differences in the various types of deep learning methods dealing with image denoising. ...
We first classify the deep convolutional neural networks (CNNs) for additive white noisy images; the deep CNNs for real noisy images; the deep CNNs for blind denoising and the deep CNNs for hybrid noisy ...
kernel idea for burst denoising Zhang et al. (2020) [255] CNN Burst denoising CNN with kernel idea and attention idea for burst denoising Zhao et al. (2019) [269] CNN Burst denoising CNN for burst denoising ...
arXiv:1912.13171v4
fatcat:4ts2xpivhreptelbgeqhljjiri
Denoising Real Bursts with Squeeze-and-excitation Residual Network
2020
IET Image Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
In this study, the authors propose a deep residual model with squeeze-and-excitation (SE) modules for the burst denoising. ...
The network contains a noise estimation convolutional neural network, which makes it capable of blind denoising. ...
Data-driven approaches are also applied to burst denoising. KPN [10] propose a CNN that predicts multiple kernels to convolve with bursts and sum to a clean estimate. Xu et al. ...
doi:10.1049/iet-ipr.2020.0041
fatcat:axtfildm4jfx7jpuejzieconf4
Dynamic Low-light Imaging with Quanta Image Sensors
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We fill the gap by proposing a student-teacher training protocol that transfers knowledge from a motion teacher and a denoising teacher to a student network. ...
QIS are single-photon image sensors with photon counting capabilities. ...
Recent reports on burst photography have focused on using deep neural networks [56] - [59] . Among these, the kernel prediction network (KPN) by Mildenhall et al. ...
arXiv:2007.08614v1
fatcat:tbtsrkvf4nf3zf7m4ejzigc2ky
Pasadena: Perceptually Aware and Stealthy Adversarial Denoise Attack
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Second, we formulate this new task as a kernel prediction problem for image filtering and propose the adversarial-denoising kernel prediction that can produce adversarial-noiseless kernels for effective ...
., whether the image denoising can be given the capability of fooling the state-of-the-art deep neural networks (DNNs) while enhancing the image quality. ...
[13] construct a UNet-based network to predict kernels for handling burst images and achieve impressive denoising performance. ...
arXiv:2007.07097v3
fatcat:3pdmqigybvay3a2mqxiz67ri2m
Non-uniform Blur Kernel Estimation via Adaptive Basis Decomposition
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2021 pre-print arXiv:2102.01026v1 fatcat:c4cvypkrjbed3bqk6c3ijrbdfq
Given a blurry image, a neural network is trained to estimate a set of image-adaptive basis motion kernels as well as the mixing coefficients at the pixel level, producing a per-pixel motion blur field ...
When applied to real motion-blurred images, a variational non-uniform blur removal method fed with the estimated blur kernels produces high-quality restored images. ...
Kernel Prediction Networks Recently, Kernel Prediction Networks (KPN) have been proposed for low-level vision tasks such burst denoising [31, 48] , optical flow estimation, frame interpolation [34, 35 ...
arXiv:2102.01026v2
fatcat:svnhprizjnfr3jioflfsg3n6nm
« Previous
Showing results 1 — 15 out of 576 results