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

TL;DR

PyNET-QxQ is a lightweight, efficient deep learning model tailored for QxQ Bayer pattern demosaicing in CMOS sensors, achieving high-quality image reconstruction with minimal parameters through progressive distillation.

Contribution

The paper introduces PyNET-QxQ, a novel, compact demosaicing model for QxQ Bayer CFA, and proposes a progressive distillation training method to maintain performance with fewer parameters.

Findings

PyNET-QxQ outperforms conventional algorithms in texture and edge reconstruction.

The model uses less than 2.5% of PyNET's parameters while preserving quality.

Progressive distillation effectively trains the lightweight network.

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 QxQ 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-QxQ, a lightweight demosaicing model specifically designed for QxQ 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-QxQ contains less than 2.5% of the parameters of the original PyNET while preserving its performance. Experiments using QxQ images captured by a proto type QxQ camera sensor show that PyNET-QxQ outperforms existing conventional algorithms in terms of texture and edge reconstruction, despite its significantly reduced parameter count.