FFDNet: Toward a Fast and Flexible Solution for CNN based Image Denoising

11 Oct 2017Kai ZhangWangmeng ZuoLei Zhang

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... (read more)

PDF Abstract

Evaluation results from the paper

Task Dataset Model Metric name Metric value Global rank Compare
Image Denoising BSD68 sigma15 FFDNet PSNR 31.63 # 4
Image Denoising BSD68 sigma25 FFDNet PSNR 29.19 # 4
Image Denoising BSD68 sigma50 FFDNet PSNR 26.29 # 5