Learning a Convolutional Neural Network for Non-uniform Motion Blur Removal

Abstract

In this paper, we address the problem of estimating and removing non-uniform motion blur from a single blurry image. We propose a deep learning approach to predicting the probabilistic distribution of motion blur at the patch level using a convolutional neural network (CNN). We further extend the candidate set of motion kernels predicted by the CNN using carefully designed image rotations. A Markov random field model is then used to infer a dense non-uniform motion blur field enforcing motion smoothness. Finally, motion blur is removed by a non-uniform deblurring model using patch-level image prior. Experimental evaluations show that our approach can effectively estimate and remove complex non-uniform motion blur that is not handled well by previous approaches.

This is a final version accepted by CVPR 2015

Cited by in corpus (33)

• Learning-Based View Synthesis for Light Field Cameras
• DeepNIS: Deep Neural Network for Nonlinear Electromagnetic Inverse Scattering
• Compression Artifacts Removal Using Convolutional Neural Networks
• DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better
• Review of data analysis in vision inspection of power lines with an in-depth discussion of deep learning technology
• Video Frame Interpolation via Adaptive Separable Convolution
• A Neural Approach to Blind Motion Deblurring
• Deep Learning Convolutional Networks for Multiphoton Microscopy Vasculature Segmentation
• Deep Video Deblurring
• The World of Fast Moving Objects
• Video Frame Interpolation via Adaptive Convolution
• Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring
• Crafting a Toolchain for Image Restoration by Deep Reinforcement Learning
• Online Video Deblurring via Dynamic Temporal Blending Network
• Learning a Discriminative Prior for Blind Image Deblurring
• Scale-recurrent Network for Deep Image Deblurring
• Learning to Extract a Video Sequence from a Single Motion-Blurred Image
• Bringing a Blurry Frame Alive at High Frame-Rate with an Event Camera
• DeepDeblur: Fast one-step blurry face images restoration
• End-to-End Learning for Image Burst Deblurring
• Simultaneous Stereo Video Deblurring and Scene Flow Estimation
• Deep Network Interpolation for Continuous Imagery Effect Transition
• Robust Blind Deconvolution via Mirror Descent
• Lensless computational imaging through deep learning
• What Is Around The Camera?
• Learning degraded image classification with restoration data fidelity
• Human-Aware Motion Deblurring
• cvpaper.challenge in 2016: Futuristic Computer Vision through 1,600 Papers Survey
• A Deep Optimization Approach for Image Deconvolution
• Learning Collaborative Generation Correction Modules for Blind Image Deblurring and Beyond
• Fiber Orientation Estimation Guided by a Deep Network
• Deep Learning for Cornea Microscopy Blind Deblurring
• cvpaper.challenge in 2015 - A review of CVPR2015 and DeepSurvey