Status: To Do
Fix Version/s: None
Team:Data Release Production
(I thought this ticket existed, but I can't find it).
In production we will almost certainly want to use a pre-defined set of stars to determine the PSF for each visit. Advantages include:
- The ability to use deeper data to reject compact galaxies
- The ability to use other bands (e.g. NIR) to reject compact galaxies
- Less stochasticity in PSF estimation
- More robust in the face of bad data (e.g. clouds)
It's a little tricky as we need decent astrometry before running the PSF estimation code.