artswan:
your idea sound intriguing...yeah, make a little experiment and see what happens.
However, you may need a calibrated temeprature/humidity sensor alongside when you are doing that to compare and correlate the ADC values with actual %RH readings.
I know for instance, that most humidity sensors have a substantial temperature coefficient, which must be used to correct the raw reading. Additionally, the results most likely would be non-linear, and you will require a lookup table to convert them to meaningful %RH values.
Lastly, I suspect that the capacitive readings could also have some supply voltage dependency, just like a normal ADC readings. As such, I would ensure when you perform the experiments that the supply voltage is accurate and recorded...of course the best would be to use a variable supply to adjust it to exactly +5.00 volts and eliminate that variable.
All in all, this can be a very nice experiment to perform. Even though the conclusion could be that the idea is not useable, you will have practiced and learned a lot of new concepts.
Have it a go!