Limitations of Data-Driven Spectral Reconstruction: An Optics-Aware Analysis

Hyperspectral imaging empowers machine vision systems with the distinct capability of identifying materials through recording their spectral signatures. Recent efforts in data-driven spectral reconstruction aim at extracting spectral information from RGB images captured by cost-effective RGB cameras, instead of dedicated hardware. Published work reports exceedingly high numerical scores for this reconstruction task, yet real-world performance lags substantially behind. In this paper we systematically analyze the performance of such methods with three groups of dedicated experiments. First, we evaluate the practical overfitting limitations with respect to current datasets by training the networks with less data, validating the trained models with unseen yet slightly modified data, and cross-dataset validation. Second, we reveal fundamental limitations in the ability of RGB to spectral methods to deal with metameric or near-metameric conditions, which have so far gone largely unnoticed due to the insufficiencies of existing datasets. We achieve this by validating the trained models with metamer data generated by metameric black theory and re-training the networks with various forms of metamers. This methodology can also be used for data augmentation as a partial mitigation of the dataset issues, although the RGB to spectral inverse problem remains fundamentally ill-posed. Finally, we analyze the potential for modifying the problem setting to achieve better performance by exploiting some form of optical encoding provided by either incidental optical aberrations or some form of deliberate optical design. Our experiments show that such approaches do indeed provide improved results under certain circumstances, however their overall performance is limited by the same dataset issues as in the plain RGB to spectral scenario. We therefore conclude that future progress on snapshot spectral imaging will heavily depend on the generation of improved datasets which can then be used to design effective optical encoding strategies.