NeRD: Neural field-based Demosaicking
This work addresses demosaicking for image processing, presenting a novel method that improves performance but appears incremental relative to existing neural approaches.
The paper tackled the problem of generating full-color images from Bayer patterns by introducing NeRD, a neural field-based demosaicking method that outperforms traditional and CNN-based approaches and significantly closes the gap to transformer-based methods.
We introduce NeRD, a new demosaicking method for generating full-color images from Bayer patterns. Our approach leverages advancements in neural fields to perform demosaicking by representing an image as a coordinate-based neural network with sine activation functions. The inputs to the network are spatial coordinates and a low-resolution Bayer pattern, while the outputs are the corresponding RGB values. An encoder network, which is a blend of ResNet and U-net, enhances the implicit neural representation of the image to improve its quality and ensure spatial consistency through prior learning. Our experimental results demonstrate that NeRD outperforms traditional and state-of-the-art CNN-based methods and significantly closes the gap to transformer-based methods.