Why do the exhaust flames from cryogenic stage engines appear to be separated from the nozzle?
Its because of black body radiation as the temperature increases the maximum light wavelength gets shifted to ultra violet (near blue)regions which humans cannot see.
This article says
According to Wikipedia
They might appear as the flame is detached from the nozzle, but that's in essence merely an illusion and the burn is there, all over the plume of the nozzle exhaust. It is however nearly translucent due to high purity of cryogenic propellants and by the chemical reaction producing molecules having high translucency. Visibility of the plume, unless you're launching at night, has actually little to do with its burn. The burn itself, as the propellant components react (fuel with the oxidizer) will be mostly in the non-visible spectrum with propellants that don't use additives / fuel impurities to shift the heat radiation to the visible spectrum. You have answered yourself why the flame appears invisible to the naked eye, but you didn't mention why it appears to be separated from the engine's nozzle, as per your question.
Both images you attached to your question, judging by the color of the exhaust plume and light refraction, use LOX/LH2 or liquid diatomic oxygen O2 and liquid diatomic hydrogen H2 in 1:2 ratio as their oxidizer + fuel propellant components, producing superheated but pure water H2O at the nozzle exhaust, and unless these plumes combine with atmospheric or propellant impurities, they will appear nearly translucent, especially given the pressure they're ejected from the nozzle injectors into the expansion chamber, where they combine. If you look closely at the images you attached, you'll however still notice a somewhat typical light refraction of water at the test facility railings behind them, more pronounced on the top photograph.
OK, so you're probably asking yourself now, what about the more white core of the exhaust plume? That is simply an area of negative pressure that forms different shapes depending on the nozzle size, shape, and of course the fuel injection rate. With some nozzles, much of this can be adjusted depending on the current performance requirements of the engine, so its position might vary, or it might even repeat itself more frequently, as for example on this image of the alcohol fueled Armadillo Aerospace's quad vehicle showing visible banding, or shock diamonds, in the exhaust plume:
In short though, the areas of negative (back-) pressure will cause this banding and additional walls forming within the plume. With molecules high refraction index, the light will scatter at these edges in a zigzag pattern (more likely much more complex with vortices forming actually) and block any direct view of the background. To rather leave this explanation to Wikipedia on Plume physics:
In short, the answer to the question is not exactly in the burn itself, since these bands, walls, and any other shapes that form within the plume as a result of a backpressure would do that just as well, if there wasn't any burn to start with. It is due to light refraction and high pressure, high velocity ejecta forming walls travelling at the opposite direction each side of the wall (detached burn), or because of the jets intersecting (diamonds).