Issue Summary

Documentation and error messages are misleading when using a release version of Caffe on Ubuntu.

Executed Command (if any)

cmake .. -DBUILD_CAFFE=OFF -DCaffe_INCLUDE_DIRS=/usr/include/caffe -DCaffe_LIBS=/usr/lib/x86_64-linux-gnu/libcaffe.so

OpenPose Output (if any)

-- The C compiler identification is GNU 7.5.0
-- The CXX compiler identificatio

File "E:\New folder\multi-person-openpose.py", line 166, in
net = cv2.dnn.readNetFromCaffe(protoFile, weightsFile)
cv2.error: OpenCV(4.2.0) C:\projects\opencv-python\opencv\modules\dnn\src\caffe\caffe_io.cpp:1132: error: (-2:Unspecified error) FAILED: fs.is_open(). Can't open "pose/coco/pose_iter_440000.caffemodel" in function 'cv::dnn::ReadProtoFromBinaryFile'

Using Gstreamer isPlaying() always returns true even when paused and using plain ofVideoPlayer it returns false when paused. It would be great to have a standard behavior.
see:
arturoc/ofxGStreamer#27

It might be useful to write the ReadME of the repo in English for ease of understanding.
It looks pretty useful.

Install OpenCV, PyQt4, Anaconda.

That means I have to install these three separately?

conda install -c https://conda.anaconda.org/menpo opencv3

Install opencv with anaconda again?

Since lollipop it's possible to record the screen: https://android.googlesource.com/platform/development/+/bf1e262/samples/ApiDemos/src/com/example/android/apis/media/projection/MediaProjectionDemo.java

Perhaps an example using a service running in the background while it captures the screen is a nice idea.

Stitching API

Opencv provide a Stitching module/API

Bascially, i would like to be able to rewrite stitching_detailed.cpp in javascript

This includes :

• Stitcher class
• Features extractors (already implemented in opencv4nodejs)
• Estimator

Description

I am trying to test dnn-detection Code
. I am receiving an error message
gocv.io/x/gocv._Cfunc_Net_BlobFromImage(0x470f010, 0x3f80101010101010,...):

I think the scalefactor parameter of the BlobFromImage seems to be wrong

The MatchTemplate() function results is given different max matching score in Windows 10.

OpenCV => 3.4.1
Operating System / Platform => Windows 10 (64 Bit)
Compiler => Visual Studio 2017

Using the same original and template images in Windows 10 the max matching score is different from Windows 7. Is there library that needs to be updated in Windows 10?

Thank you for your help,

-Ernes

lsd = cv2.createLineSegmentDetector(0)
cv2.error: OpenCV(4.1.1) /io/opencv/modules/imgproc/src/lsd.cpp:143: error: (-213:The function/feature is not implemented) Implementation has been removed due original code license issues in function 'LineSegmentDetectorImpl'

I built the library with Python3 and everything seems to be working, but I noticed that "bgs = libbgs.KNN()" (if opencv 3.x) is not working. The output of KNN is a plain white screen window. Any ideas?

(testpy3) nvidia@tegra-ubuntu:~/cviz/bgslibrary$ python Demo.py
PYTHON TYPE CONVERTERS exported
KNN()
~KNN()

System:
ubuntu 16
opencv 3.3
python3

Right now, only the facebook is showing the GIF on sharing.
We need help improving that on other social networks, please contribute!

I believe OpenCV has a phase unwrapping algorithm implemented, based on the documentation here, but I've been unable to find it to use it within the Python bindings. Does anyone know anything about this or has used this function?

• opencv-python version: 4.1.0.25