Impact of architecture on robustness and interpretability of multispectral deep neural networks
This work addresses the problem of determining optimal fusion strategies for multispectral data in vision tasks, which is incremental as it compares existing methods.
The study characterized multispectral deep learning models with different fusion approaches, quantifying their reliance on input bands and evaluating robustness to image corruptions.
Including information from additional spectral bands (e.g., near-infrared) can improve deep learning model performance for many vision-oriented tasks. There are many possible ways to incorporate this additional information into a deep learning model, but the optimal fusion strategy has not yet been determined and can vary between applications. At one extreme, known as "early fusion," additional bands are stacked as extra channels to obtain an input image with more than three channels. At the other extreme, known as "late fusion," RGB and non-RGB bands are passed through separate branches of a deep learning model and merged immediately before a final classification or segmentation layer. In this work, we characterize the performance of a suite of multispectral deep learning models with different fusion approaches, quantify their relative reliance on different input bands and evaluate their robustness to naturalistic image corruptions affecting one or more input channels.