Dmitry Budnikov
a5a21019b2
ipp_countNonZero build fix;
...
Removed IPP port for tiny arithm.cpp functions
Additional warnings fix on various platforms.
Build without OPENCL and GCC warnings fixed
Fixed warnings, trailing spaces and removed unused secure_cpy.
IPP code refactored.
IPP code path implemented as separate static functions to simplify future work with IPP code and make it more readable.
2015-06-18 12:47:07 +03:00
Vadim Pisarevsky
7dd67f4997
use _mm_sad_epu8 instead of _mm_movemask_epi8 in countNonZero. Therefore, we do not need any popcount table and this solves http://code.opencv.org/issues/4333
2015-05-14 20:33:53 +03:00
Vadim Pisarevsky
d8f360316d
fixed http://code.opencv.org/issues/3828 , added test. thanks to berak for the patch
2015-04-29 13:12:53 +03:00
Vadim Pisarevsky
ee11a2d266
fully implemented SSE and NEON cases of intrin.hpp; extended the HAL with some basic math functions
2015-04-16 23:00:26 +03:00
Maksim Shabunin
35b2139929
Basic HAL module
2015-04-07 18:16:22 +03:00
Vadim Pisarevsky
64bd77cf4d
Merge pull request #3794 from vpisarev:ubuntu_x86_fixes
2015-03-09 21:34:17 +00:00
Vadim Pisarevsky
5d8a1bf813
explained reason why IPP is disabled on 32-bit Linux. disable IPP's minMaxLoc_32f on 32-bit OSes only
2015-03-06 15:02:22 +03:00
Vadim Pisarevsky
7a52c3d9f1
1) make sure Python 2.7 is detected on Ubuntu (including libs) when it's there.
...
2) disable IPP on 32-bit Linux when OpenCV is built as shared libs. Otherwise we get linker errors
3) disable IPP's minMaxIdx 32-bit floating-point flavor in a hope that it fixes some test failures
2015-03-05 18:58:54 +03:00
Ilya Lavrenov
cf4c79ebdd
disabled some kernels for Android && NVidia
2015-03-04 14:46:49 +03:00
Ilya Lavrenov
b963d84b41
fixed bug #4215
2015-02-25 21:57:17 +03:00
Ilya Lavrenov
e9a6c5db21
sqsum
2015-01-20 21:10:06 +03:00
Ilya Lavrenov
8c94568cc3
cv::sum
2015-01-12 10:59:31 +03:00
Ilya Lavrenov
6bce6ee34a
checks
2015-01-12 10:59:31 +03:00
Ilya Lavrenov
fc0869735d
used popcnt
2015-01-12 10:59:30 +03:00
Ilya Lavrenov
5578088983
countNonZero
2015-01-12 10:59:28 +03:00
Alexander Alekhin
1c9f590f0d
IPPICV: disable NormDiff_L1_16s_C1R for IPP/ICV 8.2/8.2.1
2014-11-05 13:26:23 +03:00
Pavel Vlasov
45958eaabc
Implementation detector and selector for IPP and OpenCL;
...
IPP can be switched on and off on runtime;
Optional implementation collector was added (switched off by default in CMake). Gathers data of implementation used in functions and report this info through performance TS;
TS modifications for implementations control;
2014-10-15 14:24:41 +04:00
Ilya Lavrenov
be3efdf274
cv::sum refactoring
2014-09-30 14:36:21 +00:00
Ilya Lavrenov
1c491c42cd
fix for cornerHarris
2014-09-29 14:59:46 +00:00
Ilya Lavrenov
f50f0ba63e
cv::norm
2014-09-28 07:28:33 -07:00
Ilya Lavrenov
44ea50f1c4
cv::countNonZero
2014-09-28 07:06:53 -07:00
Ilya Lavrenov
27b933ba5a
Neon optimization of cv::sum
2014-09-22 09:22:03 +00:00
Vadim Pisarevsky
64a53de27d
Merge pull request #3185 from ElenaGvozdeva:ocl_norm
2014-09-04 08:53:47 +00:00
Elena Gvozdeva
9fe11db7e2
disabled IPP acceleration for 3-channel norms and for CV_8S only for APPLE
2014-09-04 10:38:45 +04:00
Alexander Karsakov
f57a4bf87b
Disabled minMaxIdx for 32FC1 since it occasionally fails on AMD devices (e.g. A10-6800K)
2014-09-03 14:36:51 +04:00
Poly Takahiro Horikawa
cb0ab87646
Fix arithmetic exception when pass an empty mat with IPP option on.
2014-08-19 19:19:53 -07:00
Alexander Alekhin
52ac61d87c
Merge pull request #3088 from vbystricky:ocl_enableNormEtc
2014-08-14 14:34:40 +00:00
vbystricky
942ff5be57
Disable OpenCL version of minMaxIdx with mask on AMD devices
2014-08-14 17:39:41 +04:00
vbystricky
aab6f6c56f
Set sync=true in ocl_minMaxIdx kernel call. For tests
2014-08-14 12:14:56 +04:00
vbystricky
1fe403f461
Enable OpenCL version of norm and convertScaleAbs or 32F data
...
Fix error in minmaxloc.cl
Change test for convertScaleAbs
Fix minMaxIdx for _src2 align
Change epsilon on the tests
2014-08-13 18:33:01 +04:00
Adil Ibragimov
8a4a1bb018
Several type of formal refactoring:
...
1. someMatrix.data -> someMatrix.prt()
2. someMatrix.data + someMatrix.step * lineIndex -> someMatrix.ptr( lineIndex )
3. (SomeType*) someMatrix.data -> someMatrix.ptr<SomeType>()
4. someMatrix.data -> !someMatrix.empty() ( or !someMatrix.data -> someMatrix.empty() ) in logical expressions
2014-08-13 15:21:35 +04:00
vbystricky
dfe5fff2f5
Restore OpenCL branch for minmaxLoc on float data
2014-08-11 15:57:29 +04:00
Adil Ibragimov
98d5731ad8
some formal changes (generally adding constness)
2014-08-07 15:49:14 +04:00
Alexander Alekhin
55188fe991
world fix
2014-08-05 20:12:35 +04:00
vbystricky
774d277c1f
Fix error in OpenCl version of meanstddev for continues src and not continues mask
2014-08-05 17:30:06 +04:00
Vadim Pisarevsky
a05ce00a65
Merge pull request #2879 from alalek:ocl_runtime_flags
2014-08-04 16:43:01 +00:00
vbystricky
b48e487d53
Enable ocl version of minmaxloc with mask
2014-07-31 19:12:10 +04:00
Vadim Pisarevsky
b0cd822924
trying to make the tests pass
2014-07-31 14:24:21 +04:00
Vadim Pisarevsky
5267ed46c7
1. disabled OpenCL acceleration for cv::multiply() (CV_32F), cv::divide (CV_32F), cv::convertScaleAbs (CV_32F) and cv::reduce (SUM, CV_32F), cv::reduce (MIN & MAX), cv::flip (3-channel case).
...
2. changed the number of test loops from 1 to 30 (except for cv::pow() test, which fails for yet unknown reason)
3. disabled IPP acceleration for 3-channel norms.
4. modified relativeNorm test function to handle very small values
2014-07-30 18:19:47 +04:00
Vadim Pisarevsky
af83884dc7
disabled ocl_minmaxloc on 32-bit data completely
2014-07-28 19:35:06 +04:00
Vadim Pisarevsky
1925fab427
disabled some more stuff
2014-07-28 19:32:06 +04:00
Vadim Pisarevsky
97f6dd1820
trying to make OpenCL builder green
2014-07-28 17:16:02 +04:00
Vadim Pisarevsky
f42bb0bd65
disable OpenCL difference norms with mask, since some of the tests fail
2014-07-28 15:45:54 +04:00
Elena Gvozdeva
27d97aa231
fixed
2014-07-28 12:21:00 +04:00
Elena Gvozdeva
fe29af2e58
Fixed bug in reduce.cl
2014-07-25 14:51:30 +04:00
vbystricky
690c932d7f
Change groups count for Intel
2014-07-03 15:50:56 +04:00
Ilya Lavrenov
0528d2e2b3
added 32s to 32u conversion
2014-06-23 13:09:54 +04:00
Alexander Alekhin
d9daa0cbaa
ocl: added runtime flags for debugging
2014-06-19 17:52:46 +04:00
Ilya Lavrenov
316c044e06
used abs in reduction operations
2014-06-11 18:54:43 +04:00
Ilya Lavrenov
634da9f3bf
added norm_inf support to minmaxloc kernel
2014-06-09 20:05:26 +04:00
