zyuwang's profile - activity

I use a ubuntu 1604 python 3.6 docker image to set up the build environments. And successfully built opencv python3 with -DBUILD_SHARED_LIBS=OFF.

But when I copy the cv2.cpython-36m-x86_64-linux-gnu.so file to other container base on a same image, I still need to set up environments to use this library such as libjpeg, libpng, libwebp and so on.

How can I get the same result with the case directly use "pip install opencv-python"? When I use pip install opencv-python, I don't need to set up others environments any more.

Hi, berak I still have a little problem here. I have added all the -DBUILD_xxx parameters which seem related ON now, but

if necessary, here is the link of my cascade.xml file. link text

CascadeClassifier doesn't work with model.xml trained by myself. It always blocked when use method detectMultiScale() I

CascadeClassifier doesn't work with model.xml trained by myself. It always blocked when use method detectMultiScale() I

CascadeClassifier doesn't work with model.xml trained by myself. It always blocked when use method detectMultiScale() I

CascadeClassifier trained by myself blocked while I use detectMultiScale(). I have built opencv from source detached at

CascadeClassifier trained by myself blocked while I use detectMultiScale(). I have built opencv from source detached at

CascadeClassifier trained by myself blocked while I use detectMultiScale(). I have built opencv from source detached at

CascadeClassifier trained by myself blocked while I use detectMultiScale(). I have built opencv from source detached at

CascadeClassifier trained by myself blocked while I use detectMultiScale(). I have built opencv from source detached at

Thanks for your answer. I make it right now.

after build opencv with -DBUILD_SHARED_LIBS=OFF, I still need external libraries to run apps with opencv. I use a ubuntu