mlsanityData to 7 Oct 2026

Deblurring Photographs of Characters Using Deep Neural Networks

arXiv 2205.15053 ↗First posted 30 May 2022cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

In this paper, we present our approach for the Helsinki Deblur Challenge (HDC2021). The task of this challenge is to deblur images of characters without knowing the point spread function (PSF). The organizers provided a dataset of pairs of sharp and blurred images. Our method consists of three steps: First, we estimate a warping transformation of the images to align the sharp images with the blurred ones. Next, we estimate the PSF using a quasi-Newton method. The estimated PSF allows to generate additional pairs of sharp and blurred images. Finally, we train a deep convolutional neural network to reconstruct the sharp images from the blurred images. Our method is able to successfully reconstruct images from the first 10 stages of the HDC 2021 data. Our code is available at https://github.com/hhu-machine-learning/hdc2021-psfnn.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 16 of its 25 references that are in the map. First posted on arXiv 30 May 2022.

Builds on

16 of 25 references.

References · 16

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  5. 2016Pyramid Scene Parsing Networkcited by 2,472
  6. 2018Group Normalizationcited by 1,514
  7. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 902
  8. 2021Multi-Stage Progressive Image Restorationcited by 813
  9. 2016Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurringcited by 722
  10. 2017FFDNet: Toward a Fast and Flexible Solution for CNN based Image Denoisingcited by 509
  11. 2017DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networkscited by 366
  12. 2019DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Bettercited by 361
4 more
  1. 2015Learning a Convolutional Neural Network for Non-uniform Motion Blur Removalcited by 148
  2. 2019Indices Matter: Learning to Index for Deep Image Mattingcited by 93
  3. 2014Learning to Deblurcited by 66
  4. 2021Helsinki Deblur Challenge 2021: description of photographic datacited by 1

Built on by

Cited by year

Citing papers per year: 2022: 1.

1 paper.

  1. 2022Explainable bilevel optimization: an application to the Helsinki deblur challengecited by 0