Merge pull request #16556 from ChipKerchner:vectorizeIntegralSumPixels

Browse Source

* Vectorize calculating integral for line for single and multiple channels

* Single vector processing for 4-channels - 25-30% faster

* Single vector processing for 4-channels - 25-30% faster

* Fixed AVX512 code for 4 channels

* Disable 3 channel 8UC1 to 32S for SSE2 and SSE3 (slower).  Use new version of 8UC1 to 64F for AVX512.

...

This commit is contained in:

Chip Kerchner 2020-02-28 11:34:06 -05:00 committed by GitHub

parent 7ffab23a53

commit 8c24af66bd

No known key found for this signature in database

GPG Key ID: 4AEE18F83AFDEB23

1 changed files with 958 additions and 99 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1057

modules/imgproc/src/sumpixels.simd.hpp

View File

File diff suppressed because it is too large Load Diff