Range of values qs should take

[dakdouky]

Hi,

Thanks for sharing your work.
I wonder if there is an inherent range of values that the quality score would take, I see values > 2 and that looked out of the expected range of [0, 1] in the literature.

From the CR equation, for CCS [-1, 1] and NNCCS [-0.9, 1], CR looks to have a minimum value just below 0 and its maximum value lies at around 10.

Would you please elaborate on that?

Thanks

[fdbtrs]

Hi,
There is no restriction on the FIQ values range and they do not have to be within range of [0,1].
Not all approaches in the literature have FIQ value range between 0 and 1.
Some of the approaches in the literature normalized the FIQ values to have minimum of 0 and maximum of 1. Other approaches that used for example MSE loss have value range between 0 and 1 by design. However, this d has no effect on FIQ evaluation.

[FrancisDacian]

Hi, There is no restriction on the FIQ values range and they do not have to be within range of [0,1]. Not all approaches in the literature have FIQ value range between 0 and 1. Some of the approaches in the literature normalized the FIQ values to have minimum of 0 and maximum of 1. Other approaches that used for example MSE loss have value range between 0 and 1 by design. However, this d has no effect on FIQ evaluation.

in that case, how would we evaluate the value is good or not, if there a range that we can use for normalization?