Hi there, Pythran just released its 0.8.6 version. A quick reminder: Pythran is a compiler for scientific Python, it can be used to turn Python kernels as: #pythran export _brief_loop(float64[:,:], uint8[:,:], intp[:,:], int[:,:], int[:,:]) def _brief_loop(image, descriptors, keypoints, pos0, pos1): for k in range(len(keypoints)): kr, kc = keypoints[k] for p in range(len(pos0)): pr0, pc0 = pos0[p] pr1, pc1 = pos1[p] descriptors[k,p] = image[kr + pr0, kc + pc0] < image[kr + pr1, kc + pc1] In a native library that runs much faster. It's available on pypi: https://pypi.org/project/pythran/ on conda-forge (linux and osx): https://anaconda.org/conda-forge/pythran and on github: https://github.com/serge-sans-paille/pythran As a fun fact, Pythran is now used as a mean to turn Python code into pure native library used in C++ code base, without Python involved. Some people even use SWIG to turn this code into native libraries called from... Java through JNI :-) Enjoy, Serge