Vadim Pisarevsky
8b3d6603d5
another round of dnn optimization ( #9011 )
...
* another round of dnn optimization:
* increased malloc alignment across OpenCV from 16 to 64 bytes to make it AVX2 and even AVX-512 friendly
* improved SIMD optimization of pooling layer, optimized average pooling
* cleaned up convolution layer implementation
* made activation layer "attacheable" to all other layers, including fully connected and addition layer.
* fixed bug in the fusion algorithm: "LayerData::consumers" should not be cleared, because it desctibes the topology.
* greatly optimized permutation layer, which improved SSD performance
* parallelized element-wise binary/ternary/... ops (sum, prod, max)
* also, added missing copyrights to many of the layer implementation files
* temporarily disabled (again) the check for intermediate blobs consistency; fixed warnings from various builders
2017-06-28 11:15:22 +03:00
Maksim Shabunin
32d4af36e2
Fixing some static analysis issues
2017-06-27 14:30:26 +03:00
Alexander Alekhin
006966e629
trace: initial support for code trace
2017-06-26 17:07:13 +03:00
catree
a084501ecf
Add a note to morphologyEx documentation to clarify the behavior when iterations > 1.
2017-06-25 00:16:16 +02:00
Maksim Shabunin
68d01972fe
Merge pull request #8883 from abratchik:DNN.java.wrappers.fix
2017-06-20 09:05:53 +00:00
abratchik
037d8fbdcd
Refactor OpenCV Java Wrapping
2017-06-15 20:35:12 +04:00
Alexander Alekhin
b21b694444
Merge pull request #8924 from tomoaki0705:fixWarningResize
2017-06-15 16:11:18 +00:00
Tomoaki Teshima
5ad3ddc1b6
suppress warning
...
- check if compiler is Intel compiler
- remove not referenced variables
2017-06-15 18:59:48 +09:00
Alexander Alekhin
9b902adea6
Merge pull request #8832 from terfendail:perf_accumulate_fix
2017-06-14 18:49:09 +00:00
Vadim Pisarevsky
e72e34ec36
Merge pull request #8843 from terfendail:resizenn_patch
2017-06-13 17:29:30 +00:00
Vitaly Tuzov
3a5e036feb
Updated fix for accumulate performance test in case of multiple iterations
2017-06-13 19:23:34 +03:00
Vitaly Tuzov
2de1aac665
Updated alignment declarations to CV_DECL_ALIGNED macro
2017-06-13 18:44:38 +03:00
Vitaly Tuzov
59373a1ae1
AVX and SSE optimizations for resize NN
2017-06-01 19:08:55 +03:00
Woody Chow
f743603b0a
Fallback to single threaded version of IPP gaussian blur / bilateral filter when the mutlithreaded version cannot be called.
2017-06-01 13:34:50 +09:00
Woody Chow
d22fb5f949
Multithread IPP gaussian blur
2017-05-31 18:16:47 +09:00
Vadim Pisarevsky
ee257ffe9e
Merge pull request #8455 from terfendail:ovxhal_skipsmall
2017-05-26 12:10:18 +00:00
Vitaly Tuzov
1d62a025b3
Moved size restrictions for OpenVX processed images to corresponding cpp files
2017-05-25 19:25:17 +03:00
Vadim Pisarevsky
6473018d69
eliminated trailing whitespaces
2017-05-24 16:54:12 +03:00
Vadim Pisarevsky
affb60093d
Merge branch 'master' of https://github.com/MicheleCancilla/opencv into parallel_ccomp
2017-05-24 16:51:18 +03:00
mschoeneck
4a4d94f266
Merge pull request #8694 from mschoeneck:Canny
...
Parallelize Canny with custom gradient (#8694 )
* New Canny implementation. Restructuring code in parallelCanny class. Align mag buffer and map.
* Fix warnings.
* Missing SIMD check added.
* Replaced local trailingZeros in contours.cpp. Use alignSize in canny.cpp
* Fix warnings in alignSize and allocate just minimum extra columns.
* Fix another warning in map.create.
* Exchange for loop by do loop to avoid double check at the beginning.
Define extra SIMD CANNY_CHECK to avoid unnecessary continue.
2017-05-24 16:20:25 +03:00
Vadim Pisarevsky
2e056fbe8a
Merge pull request #6854 from sturkmen72:patch-8
2017-05-24 12:45:00 +00:00
Vadim Pisarevsky
9734ee13e5
Merge pull request #7865 from LaurentBerger:UserColormap
2017-05-24 12:43:55 +00:00
Vadim Pisarevsky
057c10baac
Merge pull request #8377 from ottogin:interpMultichannelImg
2017-05-24 12:38:41 +00:00
Vadim Pisarevsky
19464a3ed8
Merge pull request #8780 from vpisarev:fix_boxfilter
2017-05-23 21:46:13 +00:00
Vadim Pisarevsky
883d925f59
replaced SSE2 code with universal intrinsics; improved accuracy of the box filter; it should now be bit-exact
2017-05-23 20:04:35 +03:00
Vadim Pisarevsky
a065e4b9aa
Merge pull request #8769 from mshabunin:kw-fixes
2017-05-23 14:59:36 +00:00
Vadim Pisarevsky
55ee8b2917
Merge pull request #8182 from chacha21:drawing_performance
2017-05-23 14:53:12 +00:00
Alexander Alekhin
c5e9d1adae
macro for static analysis tools
2017-05-23 12:35:31 +03:00
chacha21
7763a86634
restored memset optimization
...
when dropping optimizations in the last commit, I forgot to keep the
simplest case where a single memset can be called
2017-05-19 16:05:00 +02:00
Vadim Pisarevsky
913a2dbdaa
Merge pull request #8399 from woodychow:filter_avx2
2017-05-17 15:06:24 +00:00
Vadim Pisarevsky
03aa69da99
Merge pull request #8697 from sovrasov:cvt_col_bgra_bgra_fix
2017-05-17 14:11:51 +00:00
Vitaly Tuzov
01f773b803
Fix for accumulate performance test in case of multiple iterations
2017-05-16 18:28:13 +03:00
Woody Chow
67fe820c75
Merging master to filter_avx2, and resolving conflicts
2017-05-16 15:34:11 +09:00
Vladislav Sovrasov
bfc4eb31cb
imgproc: fix BGRA2BGRA conversion
2017-05-15 10:23:20 +03:00
chacha21
fa4fd48072
Drop best optimizations to reduce code size
...
Only keep the ICV_HLINE_X optimization to reduce code size.
2017-05-10 13:55:39 +02:00
Vadim Pisarevsky
833832ac91
Merge pull request #8391 from woodychow:warpAffine_avx2
2017-05-03 15:09:54 +00:00
Vadim Pisarevsky
e00d0528a4
Merge pull request #8397 from woodychow:initUndistortRectifyMap_avx2
2017-05-03 14:50:22 +00:00
Vadim Pisarevsky
8a16997b1e
Merge pull request #8580 from terfendail:ovx_newperftest
2017-05-03 11:01:01 +00:00
Maksim Shabunin
f71f76add9
Merge pull request #8649 from saskatchewancatch:8647
2017-05-03 09:19:29 +00:00
saskatchewancatch
bbb785e43c
Issue 8647: Updated doc for cv::matchTemplate to reflect current support for methods when mast template is supplied.
2017-04-30 20:22:47 -06:00
Michele Cancilla
9405c6d206
Improvement of array of equivalences’ upper bound + fix some wrong comments
2017-04-27 12:53:33 +02:00
Vitaly Tuzov
2492c299f3
Extended set of existing performance test to OpenVX HAL suitable execution modes
2017-04-27 12:32:29 +03:00
Alexander Alekhin
26be2402a3
Merge pull request #8629 from lupustr3:pvlasov/icv2017u2_update2
2017-04-26 10:45:37 +00:00
Pavel Vlasov
11c2ffaf1c
Update for IPP for OpenCV 2017u2 integration;
...
Updated integrations for:
cv::split
cv::merge
cv::insertChannel
cv::extractChannel
cv::Mat::convertTo - now with scaled conversions support
cv::LUT - disabled due to performance issues
Mat::copyTo
Mat::setTo
cv::flip
cv::copyMakeBorder - currently disabled
cv::polarToCart
cv::pow - ipp pow function was removed due to performance issues
cv::hal::magnitude32f/64f - disabled for <= SSE42, poor performance
cv::countNonZero
cv::minMaxIdx
cv::norm
cv::canny - new integration. Disabled for threaded;
cv::cornerHarris
cv::boxFilter
cv::bilateralFilter
cv::integral
2017-04-25 15:53:12 +03:00
Alexander Alekhin
41d55c5095
Merge pull request #8620 from saskatchewancatch:8457
2017-04-25 12:22:51 +00:00
saskatchewancatch
3fe18392ef
Updated comments for cv::ellipse and cv::ellipse2Poly to clarify some behaviour that has confused some users.
...
Amend: Delete trailing whitespace to make doc tests happy
2017-04-25 14:31:46 +03:00
Alexander Alekhin
f1c8094f5f
Merge pull request #8575 from lupustr3:pvlasov/icv2017u2_initial_update
2017-04-21 10:55:29 +00:00
Alessandro Gentilini
fc8f1890c0
Fix markdown format.
2017-04-21 10:29:13 +02:00
Pavel Vlasov
35c7216846
IPP for OpenCV 2017u2 initial enabling patch;
2017-04-20 20:26:30 +03:00
Vitaly Tuzov
4c0d833dec
Disabled vxuConvolution call for sepFilter evaluation
2017-04-11 15:57:20 +03:00
