Bus error (core dumped) about dynaslam HOT 29 OPEN

Hi! I met an error when running the code. I have tried this both with cpu and gpu.

Initialated Mask RCNN network...
Creating net instance...
Loading net parameters...
Bus error (core dumped)

I run the Check.py successfully and can run the code without [PATH TO MASK] [PATH TO OUTPUT] parameters. Could anyone help me? Thanks!!!

Hello there! Is your problem solved? I also encountered the same problem.

from dynaslam.

@fanguohao @kxhit Hello! I also encountered the same problem. My configuration is GTX 1080. The version of my opencv is 3.4.1

from dynaslam.

@BertaBescos I am sorry to bother you. Do you have any suggestions on this problem?

from dynaslam.

I have fixed this problem. Just change your opencv version to 2.4.11

from dynaslam.

@BertaBescos I can run the code with openCV2 but not OpenCV3. I find the program not fluent either with MaskRCNN or not. Waiting for your reply. Thanks!

from dynaslam.

I could not reproduce the problem. I would suggest to do as @jsdd25 proposes and change OpenCV version to 2.4.11. We will write you back if we find out what the problem is.

from dynaslam.

I also have this problem. I find the C++ can't call the MaskRCNN.py.
DynaSLAM::SegmentDynObject::SegmentDynObject(): Assertion `this->py_module != NULL' failed.
I don't know how to solve it. Can you help me?

from dynaslam.

Are you using OpenCV 2.4.11? If not, I would recommend you to use it.
We will try to fix this issue for newer versions within the next weeks.

from dynaslam.

Are you using OpenCV 2.4.11? If not, I would recommend you to use it.
We will try to fix this issue for newer versions within the next weeks.

Thanks for your answer,I had changed my OpenCV from 2.4.13 to 2.4.11 ,but it is useless, I am looking forward to your newer versions

from dynaslam.

Then the problem must be something else. It should work without problems on OpenCV 2.4.11.
Are you running it from the DynaSLAM directory path?

from dynaslam.

Then the problem must be something else. It should work without problems on OpenCV 2.4.11.
Are you running it from the DynaSLAM directory path?

I run it from DynaSLAM directory path like the ORB_SLAM2,it succeeded. when I added the parameter [PATH_TO_MASK] [PATH_TO_OUTPUT] to call the MaskRCNN, the problem has happened.

from dynaslam.

Then the problem must be something else. It should work without problems on OpenCV 2.4.11.
Are you running it from the DynaSLAM directory path?

I found the python version is the key problem. Mask_RCNN requires the Python3.4 + TensorFlow1.3 + Keras2.0.8,but your DynaSLAM is python2.7. I will try this way to build the DynaSLAM.

from dynaslam.

Any luck with this error? I have the same problem

from dynaslam.

I have meet the trouble for several days.But ,i can't solver it .my system is Ubuntu16.04+cuda8.0+cudnn6+opencv 2.4.11+orbslam2 environment.Can anyone help me ?

from dynaslam.

I got it working finally. The issue on my side was opencv 3.3. It works with opencv 2.4.10. When I debugged I found that the problem is in Conversion.cc inside the method toNDArray(const cv::Mat& m). The method has two sections for two opencv versions (2 and 3) in case of opencv 3 the PyObject returned from Mat seems to be the problem when it gets passed to GetDynSeg() of MaskRCNN.py. I am trying to fix it now.

from dynaslam.

You are too good！Can you tell me about your configuration environment in detail ?and have you run check.py？I have trouble again yesterday, maybe I have to configure the environment for this code again.

from dynaslam.

@PushyamiKaveti Can you help me ?

from dynaslam.

@BertaBescos Can you help me? When I give [PATH TO MASK] [PATH TO OUTPUT] ,It is "core dumped".

from dynaslam.

@PushyamiKaveti @BertaBescos My environment is :Ubuntu16.04 +cuda8.0+cudnn6+opencv2.4.11+python2.7+orbslam2.

from dynaslam.

I have solved this problem, thank you for your help.@PushyamiKaveti @BertaBescos

from dynaslam.

I have solved this problem, thank you for your help.@PushyamiKaveti @BertaBescos

I meet the same problem, can you help me?
My error is Bus error too, my python is 2.7, opencv version is 2.4.11,and Ican run Check.py successfully.
@qxdaaaaa

from dynaslam.

I meet the same problem, can you help me?
My error is Bus error too, my python is 2.7, opencv version is 2.4.11 and 3.4.1,tensorflow-gpu1.4.0 and I can run Check.py successfully.
@qxdaaaaa @kxhit @fanguohao

from dynaslam.

from dynaslam.

Thanks,@qxdaaaaa. I just test the masknet .cc find that the function cv::Mat SegmentDynObject::GetSegmentation(cv::Mat &image,std::string dir, std::string name){ .... PyObject* py_mask_image = PyObject_CallMethod(this->net, const_cast<char*>(this->get_dyn_seg.c_str()),"(O)",py_image); ......} doesn't working, how to solve this problem?

from dynaslam.

I meet the same problem, can you help me? My error is Bus error too, my python is 2.7, opencv version is 2.4.11 and 3.4.1,tensorflow-gpu1.4.0 and I can run Check.py successfully.

I‘d like to ask if cuda10.0 + tensorflow1.13.1 is ok? I think I have correctly installed with python2 but I can't pass check when I use terminal to run 'python check.py'. I think I need you help.....

from dynaslam.

I have solved this problem, thank you for your help.@PushyamiKaveti @BertaBescos

How did you solve the problem? @PushyamiKaveti @qxdaaaaa
My system is Python 2.7, OpenCV 3.3.0 , tensorflow-gpu=1.13.1, CUDA 10.0 and I can run Check.py successfully.

from dynaslam.

I have solved this problem, thank you for your help.@PushyamiKaveti @BertaBescos

How did you solve the problem? @PushyamiKaveti @qxdaaaaa
My system is Python 2.7, OpenCV 3.3.0 , tensorflow-gpu=1.13.1, CUDA 10.0 and I can run Check.py successfully.

Have you run successfully in cuda10?

from dynaslam.

I got it working finally. The issue on my side was opencv 3.3. It works with opencv 2.4.10. When I debugged I found that the problem is in Conversion.cc inside the method toNDArray(const cv::Mat& m). The method has two sections for two opencv versions (2 and 3) in case of opencv 3 the PyObject returned from Mat seems to be the problem when it gets passed to GetDynSeg() of MaskRCNN.py. I am trying to fix it now.

Yes,it is the problem,the Conversion.cc failed with opencv3.And I have fixed it, here is my Conversion.cc:

#include "Conversion.h"
#include

namespace DynaSLAM
{

static void init()
{
import_array();
}

static PyObject* failmsgp(const char *fmt, ...)
{
char str[1000];

va_list ap;
va_start(ap, fmt);
vsnprintf(str, sizeof(str), fmt, ap);
va_end(ap);

PyErr_SetString(PyExc_TypeError, str);
return 0;

}

using namespace cv;
//=================== ERROR HANDLING =========================================================

static int failmsg(const char *fmt, ...) {
char str[1000];

va_list ap;
va_start(ap, fmt);
vsnprintf(str, sizeof(str), fmt, ap);
va_end(ap);

PyErr_SetString(PyExc_TypeError, str);
return 0;

}

//=================== THREADING ==============================================================
class PyAllowThreads {
public:
PyAllowThreads() :
_state(PyEval_SaveThread()) {
}
~PyAllowThreads() {
PyEval_RestoreThread(_state);
}
private:
PyThreadState* _state;
};

class PyEnsureGIL {
public:
PyEnsureGIL() :
_state(PyGILState_Ensure()) {
}
~PyEnsureGIL() {
PyGILState_Release(_state);
}
private:
PyGILState_STATE _state;
};

enum {
ARG_NONE = 0, ARG_MAT = 1, ARG_SCALAR = 2
};

class NumpyAllocator:
public MatAllocator {
public:
NumpyAllocator() {
stdAllocator = Mat::getStdAllocator();
}
~NumpyAllocator() {
}

UMatData* allocate(PyObject* o, int dims, const int* sizes, int type,
		size_t* step) const {
	UMatData* u = new UMatData(this);
	u->data = u->origdata = (uchar*) PyArray_DATA((PyArrayObject*) o);
	npy_intp* _strides = PyArray_STRIDES((PyArrayObject*) o);
	for (int i = 0; i < dims - 1; i++)
		step[i] = (size_t) _strides[i];
	step[dims - 1] = CV_ELEM_SIZE(type);
	u->size = sizes[0] * step[0];
	u->userdata = o;
	return u;
}

UMatData* allocate(int dims0, const int* sizes, int type, void* data,
		size_t* step, int flags, UMatUsageFlags usageFlags) const {
	if (data != 0) {
		CV_Error(Error::StsAssert, "The data should normally be NULL!");
		// probably this is safe to do in such extreme case
		return stdAllocator->allocate(dims0, sizes, type, data, step, flags,
				usageFlags);
	}
	PyEnsureGIL gil;

	int depth = CV_MAT_DEPTH(type);
	int cn = CV_MAT_CN(type);
	const int f = (int) (sizeof(size_t) / 8);
	int typenum =
			depth == CV_8U ? NPY_UBYTE :
			depth == CV_8S ? NPY_BYTE :
			depth == CV_16U ? NPY_USHORT :
			depth == CV_16S ? NPY_SHORT :
			depth == CV_32S ? NPY_INT :
			depth == CV_32F ? NPY_FLOAT :
			depth == CV_64F ?
								NPY_DOUBLE :
								f * NPY_ULONGLONG + (f ^ 1) * NPY_UINT;
	int i, dims = dims0;
	cv::AutoBuffer<npy_intp> _sizes(dims + 1);
	for (i = 0; i < dims; i++)
		_sizes[i] = sizes[i];
	if (cn > 1)
		_sizes[dims++] = cn;
	PyObject* o = PyArray_SimpleNew(dims, _sizes, typenum);
	if (!o)
		CV_Error_(Error::StsError,
				("The numpy array of typenum=%d, ndims=%d can not be created", typenum, dims));
	return allocate(o, dims0, sizes, type, step);
}

bool allocate(UMatData* u, int accessFlags,
		UMatUsageFlags usageFlags) const {
	return stdAllocator->allocate(u, accessFlags, usageFlags);
}

void deallocate(UMatData* u) const {
	if (u) {
		PyEnsureGIL gil;
		PyObject* o = (PyObject*) u->userdata;
		Py_XDECREF(o);
		delete u;
	}
}

const MatAllocator* stdAllocator;

};

NumpyAllocator g_numpyAllocator;

NDArrayConverter::NDArrayConverter() { init(); }

void NDArrayConverter::init()
{
import_array();
}

cv::Mat NDArrayConverter::toMat(const PyObject o)
{
cv::Mat m;
bool allowND = true;
if (!PyArray_Check(o)) {
failmsg("argument is not a numpy array");
if (!m.data)
m.allocator = &g_numpyAllocator;
} else {
PyArrayObject oarr = (PyArrayObject*) o;

	bool needcopy = false, needcast = false;
	int typenum = PyArray_TYPE(oarr), new_typenum = typenum;
	int type = typenum == NPY_UBYTE ? CV_8U : typenum == NPY_BYTE ? CV_8S :
				typenum == NPY_USHORT ? CV_16U :
				typenum == NPY_SHORT ? CV_16S :
				typenum == NPY_INT ? CV_32S :
				typenum == NPY_INT32 ? CV_32S :
				typenum == NPY_FLOAT ? CV_32F :
				typenum == NPY_DOUBLE ? CV_64F : -1;

	if (type < 0) {
		if (typenum == NPY_INT64 || typenum == NPY_UINT64
				|| type == NPY_LONG) {
			needcopy = needcast = true;
			new_typenum = NPY_INT;
			type = CV_32S;
		} else {
			failmsg("Argument data type is not supported");
			m.allocator = &g_numpyAllocator;
			return m;
		}
	}

#ifndef CV_MAX_DIM
const int CV_MAX_DIM = 32;
#endif

	int ndims = PyArray_NDIM(oarr);
	if (ndims >= CV_MAX_DIM) {
		failmsg("Dimensionality of argument is too high");
		if (!m.data)
			m.allocator = &g_numpyAllocator;
		return m;
	}

	int size[CV_MAX_DIM + 1];
	size_t step[CV_MAX_DIM + 1];
	size_t elemsize = CV_ELEM_SIZE1(type);
	const npy_intp* _sizes = PyArray_DIMS(oarr);
	const npy_intp* _strides = PyArray_STRIDES(oarr);
	bool ismultichannel = ndims == 3 && _sizes[2] <= CV_CN_MAX;

	for (int i = ndims - 1; i >= 0 && !needcopy; i--) {
		// these checks handle cases of
		//  a) multi-dimensional (ndims > 2) arrays, as well as simpler 1- and 2-dimensional cases
		//  b) transposed arrays, where _strides[] elements go in non-descending order
		//  c) flipped arrays, where some of _strides[] elements are negative
		if ((i == ndims - 1 && (size_t) _strides[i] != elemsize)
				|| (i < ndims - 1 && _strides[i] < _strides[i + 1]))
			needcopy = true;
	}

	if (ismultichannel && _strides[1] != (npy_intp) elemsize * _sizes[2])
		needcopy = true;

	if (needcopy) {

		if (needcast) {
			o = PyArray_Cast(oarr, new_typenum);
			oarr = (PyArrayObject*) o;
		} else {
			oarr = PyArray_GETCONTIGUOUS(oarr);
			o = (PyObject*) oarr;
		}

		_strides = PyArray_STRIDES(oarr);
	}

	for (int i = 0; i < ndims; i++) {
		size[i] = (int) _sizes[i];
		step[i] = (size_t) _strides[i];
	}

	// handle degenerate case
	if (ndims == 0) {
		size[ndims] = 1;
		step[ndims] = elemsize;
		ndims++;
	}

	if (ismultichannel) {
		ndims--;
		type |= CV_MAKETYPE(0, size[2]);
	}

	if (ndims > 2 && !allowND) {
		failmsg("%s has more than 2 dimensions");
	} else {

		m = Mat(ndims, size, type, PyArray_DATA(oarr), step);
		m.u = g_numpyAllocator.allocate((PyObject*)o, ndims, size, type, step);
		m.addref();

		if (!needcopy) {
			Py_INCREF(o);
		}
	}
	m.allocator = &g_numpyAllocator;
}
return m;

}

PyObject* NDArrayConverter::toNDArray(const cv::Mat& m)
{
if (!m.data)
Py_RETURN_NONE;
Mat temp,
p = (Mat) &m;
if (!p->u || p->allocator != &g_numpyAllocator) {
temp.allocator = &g_numpyAllocator;
ERRWRAP2(m.copyTo(temp));
p = &temp;
}
std::cout<u->refcount<<std::endl;
std::cout<u->userdata<<std::endl;
PyObject* o = (PyObject*) p->u->userdata;
Py_INCREF(o);
return o;

}

}

from dynaslam.

@hanxiumeng
your code doesn't work either, leads to the same error
could you share the modified conversion.cc file? did you get it to work with opencv 3?

from dynaslam.