* Fixed centering issue with make_cicle_pattern and make_acircle_pattern()

* Fixed issue where asymmetric circles were not at 45 degree angles. Also fixed support for inch measurement by converting parsing to usage of floating points for page size

* Fixed copy-paste error from experimental workspace

- fixes proper initialization of non-trivial variable

- OpenCL kernel cleanup processing is asynchronous and can be called even after forced clFinish()
- buffers are released later in asynchronous mode
- silence these false positive cases for asynchronous cleanup

* add the HH4 mode
* option to display disparity with a color map
* display current settings in the title bar
* don't close app when wanting to take screenshots

reformatting

Improve initialization performance of Brisk

fix formatting

Improve initialization performance of Brisk

formatting

Improve initialization performance of Brisk

make a lookup table for ring

use cosine/sine lookup table for theta in brisk and utilize trig identity

fix ring lookup table

use cosine/sine lookup table for theta in brisk and utilize trig identity

formatting

use cosine/sine lookup table for theta in brisk and utilize trig identity

move scale radius product to ring loop to ensure it's not recomputed for each rot

revert change

move scale radius product to ring loop to ensure it's not recomputed for each rot

remove rings lookup table

move scale radius product to ring loop to ensure it's not recomputed for each rot

fix formatting of for loop

move scale radius product to ring loop to ensure it's not recomputed for each rot

use sine/cosine approximations for brisk lookup table.

add documentation for sine/cosine lookup tables

Improve initialization performance of BRISK

* 8-bit SIFT descriptors

* use clearer parameter

* update docs

* propagate type info

* overload function for avoiding ABI-break

* bugfix: some values are undefined when CV_SIMD is absent

- allow Users to handle such case
- exception will be thrown instead

* improve Mat::operator=(Scalar)

* touch

* remove trailing whitespace

* TEST: check if old code pass test or not

* remove CV_Error

* remove warning

* fix: is -> Scalar

* 1) Mat *mat -> Mat &mat 2) return bool, add output param

* add comment

* Added ReduceSum to ONNX importer

* Fix comments

* Fix Mul

in case of OPENCV_OPENCL_DEVICE=disabled

dnn: add exhaustive fusion tests, enable more eltwise fusions

* add eltwise fusion tests, enable more eltwise fusions

* merge weighted eltwise tests with eltwise tests

* Supported ReduceSum op

* Skip test

Improving DaSiamRPN tracker sample

* changed layerBlobs in dnn.cpp and added DaSiamRPN tracker

* Improving DaSiamRPN tracker sample

* Docs fix

* Removed outdated changes

* Trying to reinitialize tracker without reloading models. Worked with LaSOT-based benchmark with reinit rate=250 frames

* Trying to reverse changes

* Moving the model in the constructor

* Fixing some issues with names

* Variable name changed

* Reverse parser arguments changes

* Add a FLANN example showing how to search a query image in a dataset

* Clean: remove warning

* Replace dependency to boost::filesystem by calls to core/utils/filesystem

* Wait for escape key to exit

* Add an example of binary descriptors support

* Add program options for saving and loading the flann structure

* Fix warnings on Win64

* Fix warnings on 3.4 branch still relying on C++03

* Add ctor to img_info structure

* Comments modification

* * Demo file of FLANN moved and renamed

* Fix distances type when using binary vectors in the FLANN example

* Rename FLANN example file

* Remove dependency of the flann example to opencv_contrib's SURF.

* Remove mention of FLANN and other descriptors that aimed at giving hint on the other options

* Cleaner program options management

* Make waitKey usage minimal in FLANN example

* Fix the conditions order

* Use cv::Ptr

* hopefully, eliminated compile warnings, errors, as well as failure in one test

* * fixed a few typos
* decreased buffer size in some cases
* added more optimal im2row branch in the case of 1x1 convolutions
* tuned fastConv to reduce the number of passes over arrays

backport of commit 77b01deb