carrretero:
The thing is it only seems to have 2 bands that both look like red or brown, plus one at the end which might be golden.
Normally I would expect another band in between the reds/browns but there is no trace of it, I can´t believe the heat has wiped it completely
Maybe the missing band was white or the color of fried paint 
[I think I see a couple of such bands between the reddish bands].
Anyway, there isn't much hope of discerning original color [excessive heat can change the color, or make what, previously, were different colors, look, instead, the same], other than having access to [for instance] a forensics lab, where you can run samples through a gas chromatograph, or some such thing, to analyse the composition of the bits of paint, and do experimental resistor frying so you can compare known samples, to samples from your mystery resistor, etc.
Or, you could, possibly, collect a representative sample of resistors and fry them, and use those in comparison, and perhaps "logic out" the colors. But, I suspect there was a band in that exposed metal region -- and if so, even forensics may not help [unless comparing resistance measurements of fried resistors...]!?
But seriously...