mirror/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2024-12-27 03:14:05 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
18,556 Commits 6 Branches 127 Tags 3 GiB
347ffbb0eb
Commit Graph

213 Commits

Author SHA1 Message Date
Vadim Pisarevsky
3bbd22cb63 Merge pull request #5808 from cedrou:hdr.mergemertens.fixalgo 2015-12-16 10:36:29 +00:00
Cédric Rousseau
d19189e80f Fixed the formula to compute the well-exposedness weight map. 2015-12-15 10:30:06 +01:00
Sergey Alexandrov
08b30580b7 Do not modify input parameter in MergeDebevec 
MergeDebevec takes camera response function as an optional input
parameter. Despite being clearly marked as input, this matrix is
overwritten during processing.

This commit adds a temporary matrix to store the log response and avoid
modification of the input.
 2015-12-14 22:24:27 +01:00
Maksim Shabunin
715887fcd5 Coverity: fixed some negative argument issues 2015-12-09 14:25:08 +03:00
Maksim Shabunin
6e9d0d9a0c Visual Studio 2015 warning and test fixes 2015-10-20 12:48:37 +03:00
Vadim Pisarevsky
3c769edea0 added extra check in CalibrateDebevec to make sure the points are within the image: 
http://code.opencv.org/issues/4124
 2015-05-14 19:54:48 +03:00
Vadim Pisarevsky
3c7f7de477 avoid possible problems with zero values in tone mapping algorithms (http://code.opencv.org/issues/4020) 2015-05-14 19:32:25 +03:00
Vadim Pisarevsky
b37aaa8303 significantly improved parallel non-local means by using granularity parameter in parallel_for_ loop. Because the algorithm deals with sliding sums, it's essential that each thread has enough work to do, otherwise the algorithm gets higher theoretical complexity and thus there is no speedup comparing to 1-thread code (at best). 2015-05-14 15:39:42 +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
Alexander Alekhin
d326c77007 fix build (related to PR #3814) 2015-03-30 16:32:00 +03:00
Alexander Alekhin
32da602bf7 fix build (related to PR #3814) 2015-03-25 19:12:36 +03:00
Erik Karlsson
01d3df0d00 Added normType parameter to fastNlMeansDenoisingMulti 2015-03-24 02:01:31 +01:00
Erik Karlsson
5f8d688664 Added parameter normType to fastNlMeansDenoising 2015-03-24 00:47:11 +01:00
Erik Karlsson
9fff7896c5 Removed fastNlMeansDenoisingColored[Multi]Abs 2015-03-23 23:23:35 +01:00
Erik Karlsson
82c54104d6 Fix warnings on Win x64 2015-03-10 01:39:43 +01:00
Erik Karlsson
812edb5fdc Fixed bug, maxDist() should reurn int, not double 2015-03-10 01:34:02 +01:00
Erik Karlsson
8aa0783995 Changed parameter type of fastNlMeansDenoising[Multi][Abs] from std::vector<float> to const std::vector<float>& 2015-03-09 23:47:49 +01:00
Erik Karlsson
5a7118a88b Merging in master 2015-03-09 22:00:58 +01:00
Erik Karlsson
b471f9ee26 Fixed call to tegra::fastNlMeansDenoising 2015-03-09 18:52:25 +01:00
Erik Karlsson
a594a0677a Cleanup 2015-03-09 16:00:24 +01:00
Erik Karlsson
c44488629a Changed parameters of fastNlMeansDenoising[Multi][Abs] from float * to std::vector<float> 2015-03-09 15:52:16 +01:00
Erik Karlsson
cc8d94c6fc Addition of per-channel h-values for fastNlMeansDenoising[Multi][Abs] 2015-03-06 20:43:55 +01:00
Erik Karlsson
324fa26848 Refactoring of OpenCL implementation 2015-03-06 19:07:13 +01:00
Erik Karlsson
41ffcc27dd Added support for h = 0.0 2015-03-06 15:06:11 +01:00
Erik Karlsson
1e82a67cc4 Additional refactoring 2015-03-06 14:28:43 +01:00
Erik Karlsson
c41efe4e30 Refactoring in preparation for per-channel h-values 2015-03-05 17:50:52 +01:00
Erik Karlsson
18be52c05b Changed LUTs from IT to int 2015-03-05 13:55:06 +01:00
Erik Karlsson
305cff36e2 Changed from IT to int for distance calculation 2015-03-05 13:36:42 +01:00
Erik Karlsson
69eae13ff3 Changed pointers from sample_t * to uchar *. Rescaling psz accordingly. 2015-03-03 03:02:44 +01:00
Erik Karlsson
ae08884854 Added support for 4-component input for fastNlMeansDenoising[Multi][Abs] 2015-03-03 01:19:34 +01:00
Erik Karlsson
87760d13fb Cleanup and addition of 4-component support for ocl_fastNlMeansDenoising 2015-03-02 22:33:14 +01:00
Erik Karlsson
50bb14a0a8 Avoiding unnecessary copy by creating borders in place after RGB -> RGBA conversion 2015-03-02 15:48:00 +01:00
Erik Karlsson
73663dcdd1 Added support for 16-bit input 2015-03-02 03:29:17 +01:00
Erik Karlsson
a9ff335a89 Added OpenCL support for FastNlMeansDenoisingAbs 2015-03-01 22:21:36 +01:00
Erik Karlsson
0fdb95e195 Refactoring and addition of CV_8UC3 to ocl_fastNlMeansDenoising 2015-03-01 20:59:34 +01:00
Vladislav Vinogradov
cda6fed41f move tegra namespace out of cv to prevent conflicts 2015-02-27 12:52:11 +03:00
Vladislav Vinogradov
44e41baffe use new functions before all tegra:: calls 2015-02-26 19:34:58 +03:00
Erik Karlsson
8e7aff4486 Changed fastNlMeansDenoising and fastNlMeansDenoisingMulti back to sum of sq distances. Moved sq sum of abs distances to fastNlMeansDenoisingAbs and fastNlMeansDenoisingMultiAbs 2015-02-18 14:59:52 +01:00
Erik Karlsson
e647b7c7e8 Calculating almost_dist2weight at full size to avoid bounds checking 2015-02-17 23:08:36 +01:00
Erik Karlsson
baf266c29e Changed from sum of squared differences to sum of abs differences 2015-02-17 21:30:52 +01:00
Erik Karlsson
c339720af9 Preparation for 16-bit colored denoising. Currently not working due to cvtColor not supportint 16-bit Lab conversion. 2015-02-13 13:38:37 +01:00
Erik Karlsson
584372bbf2 Fixed bounds checking 2015-02-13 04:33:29 +01:00
Erik Karlsson
d588c717da Using WEIGHT_THRESHOLD to limit table size. Still problematic with 16-bit and big h-values. 2015-02-13 00:11:30 +01:00
Erik Karlsson
42db9e7153 Basic 16-bit implmentation of fastNlMeansDenoising. Table-based exponetiation leads to high memory footprint and loss of precision in 16-bit mode. 2015-02-12 22:14:01 +01:00
Erik Karlsson
8368fb9ea8 Additional refactoring preparing for 16-bit implementation 2015-02-12 18:45:09 +01:00
Erik Karlsson
e178294b49 Refactoring in preparation for 16-bit implementation of fastNlMeansDenoising 2015-02-12 15:23:28 +01:00
Vladislav Vinogradov
f48befc6f0 refactor CUDA part of photo module 2015-01-15 16:46:48 +03:00
Jiri Drbalek
d71e001736 fix crash when sample point out of image boundaries 2014-12-22 21:01:17 +00:00
Yan Wang
fd2d800c06 Remove unnecessary local variable "size". 
It should also be better for optimizing when compiling kernel.

Signed-off-by: Yan Wang <yan.wang@linux.intel.com>
 2014-12-16 14:25:24 +08:00
Yan Wang
a8b4c99ed6 Improve the performance of fast_nlmeans_denoising_opencl. 
1. Remove unnecessary barriers.
2. Adjust CTA_SIZE based on the following cases for Intel platform:
   a) OCL_Photo_DenoisingGrayscale.DenoisingGrayscale
   b) OCL_Photo_DenoisingColored.DenoisingColored
 2014-11-12 22:34:33 +08:00