Alexander Alekhin [Sun, 8 Jul 2018 13:50:35 +0000 (13:50 +0000)]
Merge pull request #11911 from berak:core_fix_autobuffer_opengl
berak [Sun, 8 Jul 2018 07:51:06 +0000 (09:51 +0200)]
core: fix autobuffer usage in opengl.cpp
Alexander Alekhin [Fri, 6 Jul 2018 14:07:50 +0000 (14:07 +0000)]
Merge pull request #11901 from alalek:fix_cuda_build
Vadim Pisarevsky [Fri, 6 Jul 2018 13:13:20 +0000 (13:13 +0000)]
Merge pull request #11867 from dkurt:dnn_ie_layers
Alexander Alekhin [Fri, 6 Jul 2018 12:31:25 +0000 (15:31 +0300)]
cuda: fix build
use cv::AutoBuffer::data() to get data pointer
Alexander Alekhin [Fri, 6 Jul 2018 12:01:31 +0000 (12:01 +0000)]
Merge pull request #11895 from alalek:fix_fn_types
Alexander Alekhin [Fri, 6 Jul 2018 12:00:43 +0000 (12:00 +0000)]
Merge pull request #11900 from alalek:build_warnings
Alexander Alekhin [Fri, 6 Jul 2018 09:41:59 +0000 (12:41 +0300)]
core(hal): eliminate build warnings
Dmitry Kurtaev [Wed, 27 Jun 2018 13:34:36 +0000 (16:34 +0300)]
Enable more deep learning tests
Alexander Alekhin [Thu, 5 Jul 2018 10:54:42 +0000 (13:54 +0300)]
fix function signatures, drop invalid casts
Alexander Alekhin [Thu, 5 Jul 2018 10:01:15 +0000 (10:01 +0000)]
Merge pull request #11719 from alalek:update_autobuffer_api
Alexander Alekhin [Thu, 5 Jul 2018 09:57:24 +0000 (09:57 +0000)]
Merge pull request #11890 from dkurt:keras_resize_nearest
Alexander Alekhin [Thu, 5 Jul 2018 09:56:50 +0000 (09:56 +0000)]
Merge pull request #11893 from dkurt:fix_11884
Alexander Alekhin [Sun, 10 Jun 2018 22:42:00 +0000 (22:42 +0000)]
opencv: Use cv::AutoBuffer<>::data()
Alexander Alekhin [Fri, 8 Jun 2018 05:55:57 +0000 (05:55 +0000)]
core: align cv::AutoBuffer API with std::vector/std::array
- added .data() methods
- added operator[] (int i)
- extend checks support to generic and debug-only cases
- deprecate existed operator* ()
Dmitry Kurtaev [Wed, 4 Jul 2018 12:50:39 +0000 (15:50 +0300)]
Disable fusion to output layers
Alexander Alekhin [Wed, 4 Jul 2018 11:08:21 +0000 (14:08 +0300)]
Merge tag '3.4.2-openvino'
Alexander Alekhin [Wed, 4 Jul 2018 11:08:11 +0000 (14:08 +0300)]
Merge tag '3.4.2'
Alexander Alekhin [Wed, 4 Jul 2018 11:05:47 +0000 (14:05 +0300)]
OpenCV 3.4.2
Dmitry Kurtaev [Wed, 4 Jul 2018 08:53:24 +0000 (11:53 +0300)]
Nearest neighbor resize from Keras
Vadim Pisarevsky [Wed, 4 Jul 2018 08:08:04 +0000 (08:08 +0000)]
Merge pull request #11875 from dkurt:dnn_fix_reshape
Alexander Alekhin [Tue, 3 Jul 2018 14:54:54 +0000 (14:54 +0000)]
Merge pull request #11882 from alalek:videoio_vfw_lower_priority
Alexander Alekhin [Tue, 3 Jul 2018 14:54:26 +0000 (14:54 +0000)]
Merge pull request #11881 from alalek:sample_use_camera_0
Alexander Alekhin [Tue, 3 Jul 2018 14:10:42 +0000 (14:10 +0000)]
Merge pull request #11879 from alalek:fix_msmf_32bit
Alexander Alekhin [Tue, 3 Jul 2018 12:54:45 +0000 (15:54 +0300)]
videoio(VFW): lower priority (after DSHOW)
Alexander Alekhin [Tue, 3 Jul 2018 12:44:53 +0000 (15:44 +0300)]
samples: videocapture_camera use VideoCapture with 0 index
Not all backends support -1 index.
Alexander Alekhin [Tue, 3 Jul 2018 12:28:55 +0000 (15:28 +0300)]
videoio(MSMF): fix 32-bit build crash
Alexander Alekhin [Tue, 3 Jul 2018 09:15:15 +0000 (09:15 +0000)]
Merge pull request #11876 from ilovezfs:patch-1
ilovezfs [Tue, 3 Jul 2018 06:31:39 +0000 (06:31 +0000)]
Python 3.7 compatability
The result of PyUnicode_AsUTF8() is now of type const char * rather of
char *.
Dmitry Kurtaev [Tue, 3 Jul 2018 05:26:43 +0000 (08:26 +0300)]
Add Reshape layer tests
Alexander Alekhin [Mon, 2 Jul 2018 15:02:13 +0000 (15:02 +0000)]
Merge pull request #11873 from catree:add_tutorial_features2d_java_python2
catree [Mon, 2 Jul 2018 12:59:00 +0000 (14:59 +0200)]
Add Java and Python code for AKAZE local features matching tutorial. Fix incorrect uses of Mat.mul() in Java code.
Uniform Lowe's ratio test in the code.
Alexander Alekhin [Mon, 2 Jul 2018 11:54:34 +0000 (11:54 +0000)]
Merge pull request #11869 from alalek:fix_11827
Alexander Alekhin [Mon, 2 Jul 2018 10:08:33 +0000 (10:08 +0000)]
Merge pull request #11865 from csukuangfj:patch_1
Alexander Alekhin [Mon, 2 Jul 2018 09:51:49 +0000 (12:51 +0300)]
imgcodecs(sunras): avoid buffer overrun
`src_pitch` may be large than data `step`
Kuang Fangjun [Mon, 2 Jul 2018 01:45:28 +0000 (09:45 +0800)]
remove a redundant option.
Alexander Alekhin [Sat, 30 Jun 2018 04:52:36 +0000 (04:52 +0000)]
Merge pull request #11856 from alalek:videoio_msmf_fix_check
Alexander Alekhin [Sat, 30 Jun 2018 04:51:20 +0000 (04:51 +0000)]
Merge pull request #11860 from alalek:videoio_msmf_remove_cxx11_code
Alexander Alekhin [Sat, 30 Jun 2018 04:50:30 +0000 (04:50 +0000)]
Merge pull request #11859 from alalek:fix_videoio_msmf_win7
Alexander Alekhin [Fri, 29 Jun 2018 20:34:14 +0000 (23:34 +0300)]
videoio(msmf): avoid using of C++11 code
build fails with MSVS 2012 without additional flags
Alexander Alekhin [Fri, 29 Jun 2018 19:54:58 +0000 (22:54 +0300)]
videoio(MSMF): avoid OpenCV load failure on Win7 machines
OpenCV binaries are compiled on Win10 environment
Alexander Alekhin [Fri, 29 Jun 2018 15:02:22 +0000 (15:02 +0000)]
Merge pull request #11854 from dkurt:dnn_tf_data_layouts_v2
Alexander Alekhin [Fri, 29 Jun 2018 15:01:20 +0000 (15:01 +0000)]
Merge pull request #11851 from pengli:3.4
Alexander Alekhin [Fri, 29 Jun 2018 12:23:48 +0000 (15:23 +0300)]
videoio(MSMF): remove "always true" check
Li Peng [Wed, 27 Jun 2018 07:15:56 +0000 (15:15 +0800)]
pooling ocl kernel optimization
set global size with real output size, also optimize
max pooling index computation if necessary.
Signed-off-by: Li Peng <peng.li@intel.com>
Dmitry Kurtaev [Fri, 29 Jun 2018 06:50:14 +0000 (09:50 +0300)]
Add a planar data layout tracking for TensorFlow importer
Vadim Pisarevsky [Thu, 28 Jun 2018 10:34:42 +0000 (10:34 +0000)]
Merge pull request #11852 from dkurt:dnn_dldt_ir_outs
Dmitry Kurtaev [Thu, 28 Jun 2018 06:09:11 +0000 (09:09 +0300)]
Set output layers names and types for models in DLDT's intermediate representation
Vadim Pisarevsky [Wed, 27 Jun 2018 18:17:39 +0000 (18:17 +0000)]
Merge pull request #11849 from catree:add_tutorial_imgproc_java_python4
Vadim Pisarevsky [Wed, 27 Jun 2018 18:12:10 +0000 (18:12 +0000)]
Merge pull request #11768 from alalek:videoio_msmf_async_live_capture
catree [Wed, 27 Jun 2018 16:48:32 +0000 (18:48 +0200)]
Add Java and Python code for Image Segmentation with Distance Transform and Watershed Algorithm tutorial. Use more Pythonic code.
Vadim Pisarevsky [Wed, 27 Jun 2018 14:26:27 +0000 (14:26 +0000)]
Merge pull request #11804 from mshabunin:gst-sample
Vadim Pisarevsky [Wed, 27 Jun 2018 13:47:35 +0000 (16:47 +0300)]
Filter homography decomp: updated PR #7153 (#11846)
* Add functionality to filter homography decompositions
* documentation + small refactor
* fix comparing int to size_t (compiler warning)
* fix whitespace issues
* clarification of function return values in documentation
* refactor of function parameters and change in loop nesting
* cleanup useless .h, fix size_t to int compare, small refactor
* fix documentation and whitespace
* change output from return value to outputarray parameter
* update function docs to reflect changes in parameters
* whitespace
* failing test
* fixed warnings related to extended initialisers and improper types
* initialize vectors from arrays
* initialize vectors from arrays part 2
* fix whitespace
* fix trailing whitespace
* Include <inttypes.h> in test_filter_homography_decomp.cpp, should fix 'uint8_t' : undeclared identifier error
* updated the test (made it shorter and providing better diagnostic) and significantly improved implementation (get rid of heavy repeated and/or unnecessary operations)
* fixed compile warning; removed trailing whitespace
Nesterov Alexander [Wed, 27 Jun 2018 13:37:10 +0000 (16:37 +0300)]
Init qrcode algo (#11829)
Vadim Pisarevsky [Wed, 27 Jun 2018 09:56:02 +0000 (09:56 +0000)]
Merge pull request #11839 from dkurt:dnn_ie_r2
Vadim Pisarevsky [Tue, 26 Jun 2018 19:27:25 +0000 (19:27 +0000)]
Merge pull request #11840 from dkurt:dnn_tf_nchw
branka-plateiq [Tue, 26 Jun 2018 13:40:28 +0000 (06:40 -0700)]
Pass RANSAC parameters as function input (#10569)
* Pass RANSAC parameters as function input
* Clean up unnecessary code
* Keep the original function signature
* Clean up based on PR comments
Replace array with vector.
Correct naming convention for input variables.
Add checks on input variables.
* Use vector instead of array for dynamic size
* Revert change.
* Use dynamic array
* Fix wrong syntax in array allocation
* Undo change
* Fix variable name
* Use vector and not array
* fixed compile warning on Windows
Dmitry Kurtaev [Tue, 26 Jun 2018 13:13:40 +0000 (16:13 +0300)]
Parse strides and convolution kernel shapes considering data layout
Dmitry Kurtaev [Tue, 26 Jun 2018 12:38:08 +0000 (15:38 +0300)]
Update Inference Engine tests
Vadim Pisarevsky [Tue, 26 Jun 2018 12:17:55 +0000 (12:17 +0000)]
Merge pull request #11836 from terfendail:msmf_camfallback
Vadim Pisarevsky [Tue, 26 Jun 2018 12:12:24 +0000 (12:12 +0000)]
Merge pull request #11835 from dkurt:dnn_tf_two_inputs
Vitaly Tuzov [Tue, 26 Jun 2018 11:04:02 +0000 (14:04 +0300)]
Disable MSMF VideoCapture fallback to existing camera in case provided index is out of range.
Dmitry Kurtaev [Tue, 26 Jun 2018 10:32:28 +0000 (13:32 +0300)]
Multiple inputs for TensorFlow models
Vadim Pisarevsky [Tue, 26 Jun 2018 06:36:27 +0000 (06:36 +0000)]
Merge pull request #11826 from dkurt:dnn_tf_data_layouts
Vadim Pisarevsky [Tue, 26 Jun 2018 06:35:18 +0000 (09:35 +0300)]
temporarily disabled OpenCL use in DNN module on Mac (#11828)
* temporarily disabled OpenCL use in DNN module on Mac, since some of the tests fail
* disable OpenCL in DNN on Mac at CMake level, not source level (thanks to alalek for the advice)
Dmitry Kurtaev [Mon, 25 Jun 2018 12:15:25 +0000 (15:15 +0300)]
Use layers consumers to predict data layout
Li, Peng [Mon, 25 Jun 2018 14:06:18 +0000 (22:06 +0800)]
update convolution opencl kernels in dnn module (#11762)
* optimize ocl kernel enqueue in fc layer
Signed-off-by: Li Peng <peng.li@intel.com>
* use CV_LOG_INFO in convolution auto tuning
Signed-off-by: Li Peng <peng.li@intel.com>
* update convolution IDLF kernel
extend parameter tuning range, also cleanup
ocl kernel implementation
Signed-off-by: Li Peng <peng.li@intel.com>
* update in-memory convolution cache config
fp16 and fp32 cache config are stored separately
Signed-off-by: Li Peng <peng.li@intel.com>
Maksim Shabunin [Mon, 25 Jun 2018 14:02:58 +0000 (17:02 +0300)]
cmake: function for application creation (#11736)
* apps: add Win32 friendly opencv_version app
Improve experience of launching app from explorer:
- application just flash (open/close) the console window
Suggested Win32 application flavor additionally shows MessageBox
and waits for User interaction.
* cmake: added unified application creation function
Vadim Pisarevsky [Mon, 25 Jun 2018 12:54:09 +0000 (12:54 +0000)]
Merge pull request #11743 from alalek:samples_avoid_double_VideoCapture_open
Rostislav Vasilikhin [Mon, 25 Jun 2018 10:19:20 +0000 (13:19 +0300)]
Kinect2 support added to OpenNI2 backend (#11794)
* Kinect2 support added to OpenNI2 backend
* more fixes for OpenNI2 backend
* whitespace fixed
* libfreenect2 support added
* const cast added
* mutable removed
Vadim Pisarevsky [Mon, 25 Jun 2018 10:16:27 +0000 (10:16 +0000)]
Merge pull request #11812 from dkurt:dnn_interp_layer_impl
Alexander Alekhin [Sat, 23 Jun 2018 18:51:06 +0000 (18:51 +0000)]
Merge pull request #11790 from berak:dnn_fix_coco_txt
Alexander Alekhin [Sat, 23 Jun 2018 17:55:12 +0000 (17:55 +0000)]
Merge pull request #11815 from tomoaki0705:suppressDeprecatedWarning
Tomoaki Teshima [Sat, 23 Jun 2018 12:35:25 +0000 (21:35 +0900)]
suppress warning while building the samples
Alexander Alekhin [Sat, 23 Jun 2018 08:46:29 +0000 (08:46 +0000)]
Merge pull request #11808 from tomoaki0705:caroteneSuppressWarnings
Dmitry Kurtaev [Fri, 22 Jun 2018 16:26:47 +0000 (19:26 +0300)]
Implement Interp layer using Resize layer
Tomoaki Teshima [Fri, 22 Jun 2018 11:17:22 +0000 (20:17 +0900)]
suppress warning on Arm platform
Alexander Alekhin [Fri, 22 Jun 2018 05:23:55 +0000 (05:23 +0000)]
Merge pull request #11773 from alalek:dnn_ocl_update_force_tuning_flag
Alexander Alekhin [Fri, 15 Jun 2018 13:13:52 +0000 (16:13 +0300)]
dnn(ocl): fix external / predefined builtin configuration behavior
OPENCV_OCL4DNN_FORCE_AUTO_TUNING should ignore existed configuration from:
- builtin predefined configurations (for Intel OpenCL iGPUs)
- external configuration (via OPENCV_OCL4DNN_CONFIG_PATH)
Prefer external configuration over builtin.
Alexander Alekhin [Thu, 21 Jun 2018 17:45:21 +0000 (17:45 +0000)]
Merge pull request #11791 from dkurt:dnn_shufflenet
Dmitry Kurtaev [Tue, 19 Jun 2018 11:35:07 +0000 (14:35 +0300)]
Add ShuffleChannel layer
Maksim Shabunin [Wed, 20 Jun 2018 11:15:34 +0000 (14:15 +0300)]
gstreamer sample: fixed incorrect pipelines being generated
Alexander Alekhin [Thu, 21 Jun 2018 11:51:54 +0000 (11:51 +0000)]
Merge pull request #11798 from dkurt:dnn_layers_names_with_dot
Dmitry Kurtaev [Wed, 20 Jun 2018 11:25:24 +0000 (14:25 +0300)]
Remove undocumented feature to retreive layers outputs by indices
berak [Tue, 19 Jun 2018 08:45:49 +0000 (10:45 +0200)]
dnn: add a coco labels file for yolov3
Alexander Alekhin [Sun, 17 Jun 2018 13:25:44 +0000 (13:25 +0000)]
Merge pull request #11775 from alalek:issue_11771
Alexander Alekhin [Sat, 16 Jun 2018 14:11:44 +0000 (14:11 +0000)]
Merge pull request #11774 from adamrankin:cap_dshow_autofocus
Alexander Alekhin [Fri, 15 Jun 2018 16:37:58 +0000 (19:37 +0300)]
photo: remove redundant broken check
`dest(roi_d)` operation contains similar check inside.
Adam Rankin [Thu, 14 Jun 2018 22:24:14 +0000 (18:24 -0400)]
Adding the ability to toggle autofocus on/off for DirectShow webcams
Alexander Alekhin [Thu, 14 Jun 2018 15:59:48 +0000 (15:59 +0000)]
Merge pull request #11760 from vishwesh5:patch-1
Alexander Alekhin [Wed, 13 Jun 2018 09:47:53 +0000 (12:47 +0300)]
videoio: MSMF async reader for camera stream
Synchronized reading from camera with heavy frame processing
provides bad effects (huge frame latency, processing frames from the past).
Generally, there is no way to process each frame and some frames will be dropped.
Allow preventive frame dropping to reduce lag of processed frames.
This mode is applied to cameras only (opened by 'index').
Alexander Alekhin [Thu, 14 Jun 2018 13:18:30 +0000 (13:18 +0000)]
Merge pull request #11766 from dkurt:dnn_darknet_avgpool_softmax
Alexander Alekhin [Thu, 14 Jun 2018 13:17:54 +0000 (13:17 +0000)]
Merge pull request #11763 from dkurt:dnn_keras_relu6_v2
Alexander Alekhin [Thu, 14 Jun 2018 13:01:22 +0000 (13:01 +0000)]
Merge pull request #11764 from alalek:videoio_msmf_refactor_ComPtr_wrapper
Dmitry Kurtaev [Thu, 14 Jun 2018 12:22:08 +0000 (15:22 +0300)]
Import average pooling and softmax layers from Darknet
Alexander Alekhin [Sun, 10 Jun 2018 12:32:09 +0000 (15:32 +0300)]
videoio(msmf): cleanup unused methods from ComPtr wrapper
Use Win32 TRUE/FALSE in SetUINT32() calls for bool parameters
Dmitry Kurtaev [Thu, 14 Jun 2018 10:30:30 +0000 (13:30 +0300)]
Import ClipByValue from Keras
Alexander Alekhin [Thu, 14 Jun 2018 09:57:38 +0000 (09:57 +0000)]
Merge pull request #11735 from alalek:videoio_msmf_fix_configureHW_reopen_condition
Alexander Alekhin [Thu, 14 Jun 2018 09:57:02 +0000 (09:57 +0000)]
Merge pull request #11755 from alalek:dnn_backend_default_config
vishwesh5 [Thu, 14 Jun 2018 07:00:06 +0000 (12:30 +0530)]
Corrected formula
Corrected formula for weighted within-class variance