Is that something that can be done without removing the wheel? Although I imagine it might ultimately be more difficult, because there is a much smaller working place...
not really.. in fact, the best way is wheel removal, and rotor removal, in order to correctly access the inner surface of the rotors..
I use a pneumatic orbital sander, what is called in a body shop as a 'DA' sander, it oscillates while it rotates, with 120 grit Emory or garnet paper, and evenly circle it around the plate, multiple times, cleaning with the spray, and drying it with air blasts, between.. what you are doing is removing adhered "migrated" pad material, and also micro ridges, restoring some "tooth" to the surface.. I do this whenever I replace pads, especially if i change "pad compound/brand".. I even do this on my trucks and other vehicles rotors, and even did "brake drums" by hand in ancient times... yeah, dating back to when my daddy taught me the process.... It always makes a lot of sense, so.. I just do it. I have no issues, and whenever I had some "shudder", on the c14, as it was the first bike that I ever felt it rise up on, it's always worked..
400 grit sandpaper is just too fine to make any effect, it just "polishes" the high spots, and does nothing to smooth the rotor out, or reach into the depth of the micro grooving that exists.. it really needs to get "some bite", into base material, and knock off the high points, to make any effect... mind you, you can get great results doing this, without removing any appreciable amount of the rotor's thickness.. .
as for "abrasive blasting", I have done this in the past, and have seen many people attempt it not without poor results.. biggest problem you will have is "superheating" the surface, which causes warps... it happens very rapidly.. hard to control... and doing it on a "stationary" rotor, by hand, will almost everytime result in warpage..
the methods I used, where done on a "jig" I made that would actually hold the rotor, and spin it via a drill, connected to a chain, connected to the jig, at a significant "RPM" speed.. (about 200 rpm approx), and directing the abrasive "spray" from a distance and an angle, to not overheat the rotor.. if you saw sparks, it was too close.. about 18-20" away was about right, and I did not use "sand", I used "glass bead", as sand was much more heat inducing than the glass...
it' is very touchy to do, It took me a long while to perfect the "system", but was worth the effort in the long run, as I ended up making and selling some units for commercial "metal guys" in my hometown, to utilize.. for the home mechanic, it's way over the $$$ limits to try.