Merge pull request #15274 from ChipKerchner:lkpyramidToHal
* Convert lkpyramid from SSE SIMD to HAL - 90% faster on Power (VSX). * Replace stores with reduce_sum. Rework to handle endianess correctly. * Fix compiler warnings by casting values explicitly to shorts * Switch to CV_SIMD128 compiler definition. Unroll loop to 8 elements since we've already loaded the data.
Showing
This diff is collapsed.
Please
register
or
sign in
to comment