Learning to Jointly Deblur, Demosaick and Denoise Raw Images

Thomas Eboli (1) , Jian Sun (2) , Jean Ponce (1)

1 WILLOW - Models of visual object recognition and scene understanding
2 Xjtu - Xi'an Jiaotong University

We address the problem of non-blind deblurring and demosaicking of noisy raw images. We adapt an existing learningbased approach to RGB image deblurring to handle raw images by introducing a new interpretable module that jointly demosaicks and deblurs them. We train this model on RGB images converted into raw ones following a realistic invertible camera pipeline. We demonstrate the effectiveness of this model over two-stage approaches stacking demosaicking and deblurring modules on quantitive benchmarks. We also apply our approach to remove a camera's inherent blur (its colordependent point-spread function) from real images, in essence deblurring sharp images.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Thomas Eboli, Jian Sun, Jean Ponce. Learning to Jointly Deblur, Demosaick and Denoise Raw Images. 2021. ⟨hal-03197462⟩