Dynamic Scene Deblurring With Parameter Selective Sharing and Nested Skip Connections
Authors:
- 1 June 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3843-3851
Abstract
Dynamic Scene deblurring is a challenging low-level vision task where spatially variant blur is caused by many factors, e.g., camera shake and object motion. Recent study has made significant progress. Compared with the parameter independence scheme [19] and parameter sharing scheme [33], we develop the general principle for constraining the deblurring network structure by proposing the generic and effective selective sharing scheme. Inside the subnetwork of each scale, we propose a nested skip connection structure for the nonlinear transformation modules to replace stacked convolution layers or residual blocks. Besides, we build a new large dataset of blurred/sharp image pairs towards better restoration quality. Comprehensive experimental results show that our parameter selective sharing scheme, nested skip connection structure, and the new dataset are all significant to set a new state-of-the-art in dynamic scene deblurring.Keywords
This publication has 25 references indexed in Scilit:
- Accurate Image Super-Resolution Using Very Deep Convolutional Networks2016
- FlowNet: Learning Optical Flow with Convolutional Networks2015
- Learning to DeblurIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- Learning a convolutional neural network for non-uniform motion blur removal2015
- Recurrent convolutional neural network for object recognition2015
- Unnatural L0 Sparse Representation for Natural Image Deblurring2013
- From learning models of natural image patches to whole image restoration2011
- Fast motion deblurringACM Transactions on Graphics, 2009
- High-quality motion deblurring from a single imageACM Transactions on Graphics, 2008
- Total variation blind deconvolutionIEEE Transactions on Image Processing, 1998