Fitting spatial variations in photutils #2215
Description
Issue RCAL-1349 was created on JIRA by Jonathan Eisenhamer:
The last missing feature in photutils is handling variation in the PSF and its perturbations. Conceptually this means using the stamps in different parts of the image to determine how the shape of the PSF varies spatially. We can probably get most of the way here just by cutting the image up into ~quadrants and putting together a gridded PSF routine that uses the results from the different quadrants. There’s some additional logic needed about how to handle quadrants which don’t have enough stars, and how to handle the corners where we’d like to know the PSF literally in the corners but our stars have an average position somewhat away from the corner. Jay has prescriptions here that I need to review. We can revisit this when we think that the ePSF fitting is in good shape otherwise; this feature is nice to have but not critical.