P2SDenoise[data]
removes noise from the data using self-supervised linear regression,where each volume is predicted from all other volumes.
P2SDenoise[data,mask]
does the same but only within the mask. The regression is fitted on Iterations random row sketches of 10% of the voxels and the predictions are averaged. The option Method sets the sketch and can be "Leverage" (default, leverage score sampling) or "Count" (count sketch).
is based on DOI:10.48550/arXiv.2011.01355 (Patch2Self) and Patch2Self2 (CVPR 2024).
P2SDenoise
P2SDenoise[data]
removes noise from the data using self-supervised linear regression,where each volume is predicted from all other volumes.
P2SDenoise[data,mask]
does the same but only within the mask. The regression is fitted on Iterations random row sketches of 10% of the voxels and the predictions are averaged. The option Method sets the sketch and can be "Leverage" (default, leverage score sampling) or "Count" (count sketch).
is based on DOI:10.48550/arXiv.2011.01355 (Patch2Self) and Patch2Self2 (CVPR 2024).
Details and Options
- The following options can be given:
-
Iterations 10 Iterations is an options for RegisterData, RegisterDiffusionData, and RegisterDataTransform. It specifies the number of iterations used by the registration functions. Method "Leverage" Method is an option for various algorithm-intensive functions that specifies what internal methods they should use.