Aberration-Aware Depth-From-Focus

Abstract
Computer vision methods for depth estimation usually use simple camera models with idealized optics. For modern machine learning approaches, this creates an issue when attempting to train deep networks with simulated data, especially for focus-sensitive tasks like Depth-from-Focus. In this work, we investigate the domain gap caused by off-axis aberrations that will affect the decision of the best-focused frame in a focal stack. We then explore bridging this domain gap through aberration-aware training (AAT). Our approach involves a lightweight network that models lens aberrations at different positions and focus distances, which is then integrated into the conventional network training pipeline. We evaluate the generality of network models on both synthetic and real-world data. The experimental results demonstrate that the proposed AAT scheme can improve depth estimation accuracy without fine-tuning the model for different datasets.

Citation
Yang, X., Fu, Q., Elhoseiny, M., & Heidrich, W. (2023). Aberration-Aware Depth-From-Focus. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1–11. https://doi.org/10.1109/tpami.2023.3301931

Acknowledgements
This work was supported by the King Abdullah University of Science and Technology (KAUST) individual baseline funding.

Publisher
Institute of Electrical and Electronics Engineers (IEEE)

Journal
IEEE Transactions on Pattern Analysis and Machine Intelligence

DOI
10.1109/TPAMI.2023.3301931

arXiv
2303.04654

Additional Links
https://ieeexplore.ieee.org/document/10209238/

Permanent link to this record

Version History

Now showing 1 - 2 of 2
VersionDateSummary
2*
2023-08-31 12:57:33
Published in Journal
2023-03-12 12:25:51
* Selected version