Performance analysis

asked 2018-11-02 18:14:18 -0600

tobix10 gravatar image

I would like to prepare a test which will give me an exact answer if particular CPU (ARM) is capable of doing image processing. I'd like to test basic operations on images. It will be nice to see a performance (time or fps) from a light to heavy processing on CPU and memory.

Which basic operations should I choose? Which operations will benefit from NEON feature?

edit retag flag offensive close merge delete


maybe this ?

berak gravatar imageberak ( 2018-11-03 03:36:52 -0600 )edit

Yeah, but this is a profiler. I still have to write some code. I am not an expert in image processing and OpenCV. I would like to know which basic operations are most often part of more complicated methods. I also want to know which methods have optimization for NEON so I could compare the performance with and without this feature.

tobix10 gravatar imagetobix10 ( 2018-11-04 05:13:50 -0600 )edit

imho, you have to do measurements for your use-case (which is ?)

there are ~2500 functions in the library, trying with arbitrary operations won't help you (imho)

berak gravatar imageberak ( 2018-11-04 05:20:27 -0600 )edit