Request to change the test code(or interface?)

[tegusi]

Hello! I just submitted a pr, and the robot reports errors on modules/python/test/test_facedetect.py*, as

Traceback (most recent call last):
File "/build/precommit_linux64/opencv/modules/python/test/test_facedetect.py", line 55, in test_facedetect
rects = detect(gray, cascade)
File "/build/precommit_linux64/opencv/modules/python/test/test_facedetect.py", line 18, in detect
rects[:,2:] += rects[:,:2]
TypeError: list indices must be integers or slices, not tuple

and the source file is

opencv/modules/python/test/test_facedetect.py

Lines 13 to 19 in 360758e

	def detect(img, cascade):
	rects = cascade.detectMultiScale(img, scaleFactor=1.275, minNeighbors=4, minSize=(30, 30),
	flags=cv.CASCADE_SCALE_IMAGE)
	if len(rects) == 0:
	return []
	rects[:,2:] += rects[:,:2]
	return rects

Since the C++ version of this code provides vector, the python version should return LIST OF TUPLES instead of NUMPY 2D ARRAY as above, which satisfy the consistency of vector convertion between Python and C as well.

opencv/modules/objdetect/src/cascadedetect.cpp

Lines 1351 to 1356 in 360758e

	void CascadeClassifierImpl::detectMultiScale( InputArray _image, std::vector<Rect>& objects,
	std::vector<int>& rejectLevels,
	std::vector<double>& levelWeights,
	double scaleFactor, int minNeighbors,
	int /*flags*/, Size minObjectSize, Size maxObjectSize,
	bool outputRejectLevels )

Which can be fixed by convert it to numpy array
rects = np.array(rects)

System information (version)

• OpenCV => 4.1.0
• Operating System / Platform => Ubuntu 16.04
• Compiler => gcc 5.4.0

Detailed description

Steps to reproduce

[alalek]

should return LIST OF TUPLES

Tuples are almost not usable for future processing (including math-based), so we using np.array instead of tuples whenever possible.
So we don't asking users to do this: rects = np.array(rects)

[tegusi]

However, the reverse conversion(numpy array to vector) doesn't work well(pyopencv_to). In this issue, after my final pr the original code still fails because of python binding:

terminate called after throwing an instance of 'cv::Exception' what(): OpenCV(4.1.0-pre) /home/tegusi/opencv-master/modules/core/src/copy.cpp:254: error: (-215:Assertion failed) channels() == CV_MAT_CN(dtype) in function 'copyTo'

corresponds to this:

opencv/modules/python/src2/cv2.cpp

Lines 1174 to 1176 in 066c074

	Mat m;
	pyopencv_to(obj, m, info);
	m.copyTo(value);

I've tried my best to solve this problem but failed...I wonder if you know how to solve this problem :(