SVD-DIP: Overcoming the Overfitting Problem in DIP-based CT Reconstruction

Overfitting Singular value
DOI: 10.48550/arxiv.2303.15748 Publication Date: 2023-01-01
ABSTRACT
The deep image prior (DIP) is a well-established unsupervised learning method for reconstruction; yet it far from being flawless. DIP overfits to noise if not early stopped, or optimized via regularized objective. We build on the fine-tuning of pretrained DIP, by adopting novel strategy that restricts adaptation singular values. proposed SVD-DIP uses ad hoc convolutional layers whose parameters are decomposed value decomposition. Optimizing then solely consists in values, while keeping left and right vectors fixed. thoroughly validate real-measured $\mu$CT data lotus root as well two medical datasets (LoDoPaB Mayo). report significantly improved stability optimization, overcoming overfitting noise.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES ()
CITATIONS ()
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....