For people like me, who find reading these papers verbatim hard work, one sentence in the article is worth repeating here:

"...although NW arrays used in present work have not been capable for color vision, further development of multiple nanowire types with different spectral responses may be a potentially feasible strategy to realize such a scenario in the near future. "

So the nano-wires do work, but for monochrome type vision. I think.

My experience looking at black-and-white pictures and video suggests that, compared to being blind, the difference between color vision and monochrome vision is negligible.

Squint and you'll find that vision is heavily reliant on contrast between light and dark rather than color fields.

well from there it's just mixing different wavelength ranges of "monochrome", so the difference is less than one might expect (yes I am ignoring the preprocessing cells that have to be in tune too for correct colour vision to work, but it's not that far off)