python - Fixed-length HOG descriptor for variable sized images -

python - Fixed-length HOG descriptor for variable sized images -

i'm working on object detection code, objects don't have fixed size, so;

skimage.feature.hog(obj)

doesn't give me equal length vectors(since uses fixed sized cells), , hence can't utilize learning algorithms on them.

so, tried dynamically assigning pig feature length:

from __future__ import partition def describe_object(obj, div=8): width, height = obj.shape f = skimage.feature.hog(obj, normalise=true, pixels_per_cell=(height//div, width//div)) homecoming f

but, mostly gives 2916 sized vectors, gives longer vectors (like 3402 elements long) too.

i believe happens when specific ratio between bin size , object's shape, don't know why exactly.

can help me?

you scale images fixed size, before calculating pig features.

python image-processing numpy feature-detection scikit-image