- bgfg_segm
- peopledetect
- opencv_version
- dnn/colorization
- tapi/opencl_custom_kernel
- tapi/dense_optical_flow (renamed tvl1_optical_flow)

* Use only absolute prior boxes explicit sizes. Remove scales attributes.

* Simplified PriorBox layer forward pass

- allow installing samples sources on all platforms
  even if BUILD_EXAMPLES is disabled,  fixed minor
  issues in sources installation process
- use 'example_<group>_<name>' scheme for target and binary file naming
- use single function for sample executable creation

Found via `codespell -q 3 --skip="./3rdparty" -I ../opencv-whitelist.txt`

* Intel Inference Engine deep learning backend.

* OpenFace network using Inference Engine backend

* dnn: add colorization.cpp

* Update arguments of dnn/colorization.py

* Added tool to generate text graph for SSD models from TensorFlow

* Update tf_text_graph_ssd.py

Save video file (#10502)

* Add a parameter labels to command line

* default value

* samples: caffe_googlenet.cpp minor refactoring

Signed-off-by: Li Peng <peng.li@intel.com>

* add style draw in yolo

* fix sintaxe whitespace

* fix conversion from float to int

* sample refactoring

- minor code style fixes
- avoid confusing "bottom" names
- use cv::format
- always draw object detection/roi

* Update samples

* Update calib3d.hpp

* Update calib3d.hpp

* Update calib3d.hpp

* Update calib3d.hpp

Implementation of bit-exact resize. Internal calls to linear resize updated to use bit-exact version. (#9468)

Signed-off-by: Li Peng <peng.li@intel.com>

getUMat()/getMat() calls are scope based. Results of these calls can't be
stored somewhere for future usage.