Furkan Kınlı

Furkan Kınlı

Ph.D. Candidate
Computer Science
Özyeğin University

Re: Spatial-Adaptive Network for Single Image Denoising

ML Reproducibility Challenge 2020, Accepted to ReScience Journal Publication

Abstract

In this study, we present our results and experience during replicating the paper titled “Spatial-Adaptive Network for Single Image Denoising”. This paper proposes novel spatial-adaptive denoising architecture for efficient noise removal by leveraging the deformable convolutions to adapt spatial information (i.e. edges and textures). We have implemented the model from scratch in PyTorch framework, and then have conducted real and synthetic noise experiments on the corresponding datasets. We have achieved to reproduce the results qualitatively and quantitatively.

Re-Science | OpenReview | Code

Bibtex:

@article{Mentes:2021,
  author = {Menteş, Sami and Kınlı, Furkan and Özcan, Barış and Kıraç, Furkan},
  title = ,
  journal = {ReScience C},
  year = {2021},
  month = may,
  volume = {7},
  number = {2},
  pages = ,
  doi = {10.5281/zenodo.4834672},
  url = {https://zenodo.org/record/4834672/files/article.pdf},
  code_url = {https://github.com/sami-automatic/SADNet_Replication},
  code_doi = {},
  code_swh = {swh:1:dir:1c60d43a0fe927c1f1287adefd252804c2f273b9},
  data_url = {},
  data_doi = {},
  review_url = {https://openreview.net/forum?id=yiAI9QN9nYt&noteId=SMFjCY6qG8},
  type = {Replication},
  language = {Python},
  domain = {ML Reproducibility Challenge 2020},
  keywords = {image denoising, image restoration, image processing}
}