Seite 1 von 1

Assignment 3 2A - NaN values

Verfasst: 26. Jun 2013 15:31
von ampelmann
Hi,
my extract_feature class returns some training data that contains "NaN" values. What could cause this? I guess I am doing something wrong with applying SIFT, because my harris-feature points are correct (checked visually).I apply sift on the feature points. Is there anything I might have missed? Appreciate any help! Thanks in advance :)

Re: Assignment 3 2A - NaN values

Verfasst: 26. Jun 2013 16:01
von lustiz
ampelmann hat geschrieben:Hi,
my extract_feature class returns some training data that contains "NaN" values. What could cause this? I guess I am doing something wrong with applying SIFT, because my harris-feature points are correct (checked visually).I apply sift on the feature points. Is there anything I might have missed? Appreciate any help! Thanks in advance :)
Hi, can you state where exactly NaN values are produced within your extract_features method? The pipeline consists of two steps per image: First, detecting interest points and, second, computing a SIFT feature vector per interest point (which, however, can be done in one call to sift)... So I guess it is the call to sift which gives you NaN values? Did you make sure to ignore pixels within a certain distance from the boundary?

Cheers lustiz

Re: Assignment 3 2A - NaN values

Verfasst: 26. Jun 2013 16:27
von ampelmann
Hi, thanks for that quick reply. And I really don't know what I've changed, but now it works fine... Maybe it was the coffee break.