Hyperspectral Reconstruction of Skin Through Fusion of Scattering Transform Features
This work addresses a domain-specific challenge in medical imaging by providing an incremental improvement for skin analysis applications.
The paper tackles the problem of reconstructing hyperspectral skin imagery from RGB and infrared inputs by proposing a model that matches and inverts scattering transform features, resulting in improved learning complexity and performance.
Hyperspectral imagery (HSI) is an established technique with an array of applications, but its use is limited due to both practical and technical issues associated with spectral devices. The goal of the ICASSP 2024 'Hyper-Skin' Challenge is to extract skin HSI from matching RGB images and an infrared band. To address this problem we propose a model using features of the scattering transform - a type of convolutional neural network with predefined filters. Our model matches and inverts those features, rather than the pixel values, reducing the complexity of matching while grouping similar features together, resulting in an improved learning process.