Burst Denoising with Kernel Prediction Networks
Authors:
- Ben MildenhallBMBen Mildenhall
- Jonathan T. BarronJBJonathan T. Barron
- Jiawen ChenJCJiawen Chen
- Dillon SharletDSDillon Sharlet
- Ren NgRNRen Ng
- Robert CarrollRCRobert Carroll
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2502-2510
Abstract
We present a technique for jointly denoising bursts of images taken from a handheld camera. 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 noise formation model, and an optimization guided by an annealed loss function to avoid undesirable local minima. Our model matches or outperforms the state-of-the-art across a wide range of noise levels on both real and synthetic data.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- ProxImaLACM Transactions on Graphics, 2016
- Fast burst images denoisingACM Transactions on Graphics, 2014
- FlexISPACM Transactions on Graphics, 2014
- Video Denoising, Deblocking, and Enhancement Through Separable 4-D Nonlocal Spatiotemporal TransformsIEEE Transactions on Image Processing, 2012
- Principal components for non-local means image denoising2008
- Image Denoising by Sparse 3-D Transform-Domain Collaborative FilteringIEEE Transactions on Image Processing, 2007
- A Non-Local Algorithm for Image Denoising2005
- Radiometric CCD camera calibration and noise estimationIEEE Transactions on Pattern Analysis and Machine Intelligence, 1994
- Nonlinear total variation based noise removal algorithmsPhysica D: Nonlinear Phenomena, 1992
- Scale-space and edge detection using anisotropic diffusionIEEE Transactions on Pattern Analysis and Machine Intelligence, 1990