Sim2Real in Reconstructive Spectroscopy: Deep Learning with Augmented Device-Informed Data Simulation
This work addresses the challenge of efficient spectral signal reconstruction in spectroscopy, particularly for applications where real data collection is difficult, though it appears incremental by adapting existing deep learning techniques to a specific domain.
The paper tackles the problem of reconstructing spectral signals using only device-informed simulated data for training, which suffers from distribution shifts from real data, and achieves on-par performance with state-of-the-art optimization-based methods while significantly speeding up inference.
This work proposes a deep learning (DL)-based framework, namely Sim2Real, for spectral signal reconstruction in reconstructive spectroscopy, focusing on efficient data sampling and fast inference time. The work focuses on the challenge of reconstructing real-world spectral signals under the extreme setting where only device-informed simulated data are available for training. Such device-informed simulated data are much easier to collect than real-world data but exhibit large distribution shifts from their real-world counterparts. To leverage such simulated data effectively, a hierarchical data augmentation strategy is introduced to mitigate the adverse effects of this domain shift, and a corresponding neural network for the spectral signal reconstruction with our augmented data is designed. Experiments using a real dataset measured from our spectrometer device demonstrate that Sim2Real achieves significant speed-up during the inference while attaining on-par performance with the state-of-the-art optimization-based methods.