PyNET-Q×Q: An Efficient PyNET Variant for Q×Q Bayer Pattern Demosaicing in CMOS Image Sensors

Minhyeok Cho, Haechang Lee, Hyunwoo Je, Kijeong Kim, Dongil Ryu, Albert No

Abstract: Deep learning-based image signal processor (ISP) models for mobile cameras can generate high-quality images that rival those of professional DSLR cameras. However, their computational demands often make them unsuitable for mobile settings. Additionally, modern mobile cameras employ non-Bayer color filter arrays (CFA) such as Quad Bayer, Nona Bayer, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{Q}\times \text{Q}$ </tex-math></inline-formula> Bayer to enhance image quality, yet most existing deep learning-based ISP (or demosaicing) models focus primarily on standard Bayer CFAs. In this study, we present PyNET- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{Q}\times \text{Q}$ </tex-math></inline-formula> , a lightweight demosaicing model specifically designed for <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{Q}\times \text{Q}$ </tex-math></inline-formula> Bayer CFA patterns, which is derived from the original PyNET. We also propose a knowledge distillation method called progressive distillation to train the reduced network more effectively. Consequently, PyNET- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{Q}\times \text{Q}$ </tex-math></inline-formula> contains less than 2.5% of the parameters of the original PyNET while preserving its performance. Experiments using <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{Q}\times \text{Q}$ </tex-math></inline-formula> images captured by a prototype <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{Q}\times \text{Q}$ </tex-math></inline-formula> camera sensor show that PyNET- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\text{Q}\times \text{Q}$ </tex-math></inline-formula> outperforms existing conventional algorithms in terms of texture and edge reconstruction, despite its significantly reduced parameter count. Code and partial datasets can be found at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Minhyeok01/PyNET-QxQ</uri> .