Motion Deblurring in the Wild

TL;DR

This paper introduces a novel convolutional network architecture for image deblurring in real-world scenarios, effectively handling spatially varying blur and occlusions by leveraging a realistic dataset collected from high frame rate videos.

Contribution

The paper proposes a new multi-stage CNN architecture for wild image deblurring and introduces a realistic dataset created from high frame rate videos to improve training.

Findings

Achieves state-of-the-art deblurring performance on wild images.

Effectively handles spatially varying blur and occlusions.

Realistic dataset improves model robustness.

Abstract

The task of image deblurring is a very ill-posed problem as both the image and the blur are unknown. Moreover, when pictures are taken in the wild, this task becomes even more challenging due to the blur varying spatially and the occlusions between the object. Due to the complexity of the general image model we propose a novel convolutional network architecture which directly generates the sharp image.This network is built in three stages, and exploits the benefits of pyramid schemes often used in blind deconvolution. One of the main difficulties in training such a network is to design a suitable dataset. While useful data can be obtained by synthetically blurring a collection of images, more realistic data must be collected in the wild. To obtain such data we use a high frame rate video camera and keep one frame as the sharp image and frame average as the corresponding blurred image. We show that this realistic dataset is key in achieving state-of-the-art performance and dealing with occlusions.