Is the calculation of ncc wrong in the code?

uzh-rpg / rpg_open_remode

This repository contains an implementation of REMODE (REgularized MOnocular Depth Estimation), as described in the paper.

http://rpg.ifi.uzh.ch/docs/ICRA14_Pizzoli.pdf

GNU General Public License v3.0

488 stars 187 forks source link

Is the calculation of ncc wrong in the code? #30

Closed TianQi-777 closed 6 years ago

TianQi-777 commented 6 years ago

In the epipolar_match.cu file, why is NCC calculated as const float ncc = ncc_numerator * rsqrtf(ncc_denominator + FLT_MIN); instead of const float ncc = ncc_numerator /rsqrtf(ncc_denominator + FLT_MIN); ?

TianQi-777 commented 6 years ago

Sorry, I misunderstood the rsqrt function. A * rsqrtf(B) == A / sqrtf(B)