FFDNet: Toward a fast and flexible solution for CNN-Based image denoising

Kai Zhang, Wangmeng Zuo, Lei Zhang

Research output: Journal article publicationJournal articleAcademic researchpeer-review

1480 Citations (Scopus)


Due to the fast inference and good performance, discriminative learning methods have been widely studied in image denoising. However, these methods mostly learn a specific model for each noise level, and require multiple models for denoising images with different noise levels. They also lack flexibility to deal with spatially variant noise, limiting their applications in practical denoising. To address these issues, we present a fast and flexible denoising convolutional neural network, namely FFDNet, with a tunable noise level map as the input. The proposed FFDNet works on downsampled sub-images, achieving a good trade-off between inference speed and denoising performance. In contrast to the existing discriminative denoisers, FFDNet enjoys several desirable properties, including: 1) the ability to handle a wide range of noise levels (i.e., [0, 75]) effectively with a single network; 2) the ability to remove spatially variant noise by specifying a non-uniform noise level map; and 3) faster speed than benchmark BM3D even on CPU without sacrificing denoising performance. Extensive experiments on synthetic and real noisy images are conducted to evaluate FFDNet in comparison with state-of-the-art denoisers. The results show that FFDNet is effective and efficient, making it highly attractive for practical denoising applications.

Original languageEnglish
Pages (from-to)4608-4622
Number of pages15
JournalIEEE Transactions on Image Processing
Issue number9
Publication statusPublished - 1 Sept 2018


  • convolutional neural networks
  • Gaussian noise
  • Image denoising
  • spatially variant noise

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design


Dive into the research topics of 'FFDNet: Toward a fast and flexible solution for CNN-Based image denoising'. Together they form a unique fingerprint.

Cite this