2021/03/18 by Maximilian Schambach, Jiayang Shi, Schambach, Maximilian +3
Computer Science · Engineering · Physics and Astronomy · #Advanced Optical Sensing Technologies #Advanced Vision and Imaging #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Optical Coherence Tomography Applications
paper · pdf · doi:10.48550/arxiv.2103.10179
openalex publication_date 2021/03/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We present a novel method to reconstruct a spectral central view and its\naligned disparity map from spatio-spectrally coded light fields. Since we do\nnot reconstruct an intermediate full light field from the coded measurement, we\nrefer to this as principal reconstruction. The coded light fields correspond to\nthose captured by a light field camera in the unfocused design with a\nspectrally coded microlens array. In this application, the spectrally coded\nlight field camera can be interpreted as a single-shot spectral depth camera.\n We investigate several multi-task deep learning methods and propose a new\nauxiliary loss-based training strategy to enhance the reconstruction\nperformance. The results are evaluated using a synthetic as well as a new\nreal-world spectral light field dataset that we captured using a custom-built\ncamera. The results are compared to state-of-the art compressed sensing\nreconstruction and disparity estimation.\n We achieve a high reconstruction quality for both synthetic and real-world\ncoded light fields. The disparity estimation quality is on par with or even\noutperforms state-of-the-art disparity estimation from uncoded RGB light\nfields.\n