Hi,
According to the forums only single image tracking supports multiple simultaneous tracking, I was wondering is there is a workaround to be able to track two multi-targets simultaneously.
I would like for the user to have the ability to track two multi-target objects within one window.
Any help or work around would be greatly appreciated, also is there an ETA for such a feature to be supported?
Kind Regards,
W
It is not possible that your solution works.
Oh, but it is... :)
You made two mistakes.
Firstly you need to make this call exactly where I stated, right after InitTracker:
[self initTracker];
QCAR::setHint(QCAR::HINT_MAX_SIMULTANEOUS_IMAGE_TARGETS, 12);
const QCAR::TrackableResult* result=NULL;
int numResults=state.getNumTrackableResults();
NSLog(@"NumTrackableResults = %i",numResults);
To make it easier for you I have posted the remaining code for renderFrameQCAR() below.
HTH
// Did we find any trackables this frame?
if (state.getNumTrackableResults())
{
// Get the trackable:
const QCAR::TrackableResult* result=NULL;
int numResults=state.getNumTrackableResults();
NSLog(@"NumTrackableResults = %i",numResults);
// Browse results searching for the MultiTargets
for (int j=0;j<numResults;j++)
{
result = state.getTrackableResult(j);
if (result->isOfType(QCAR::MultiTargetResult::getClassType()))
{
QCAR::Matrix44F modelViewMatrix =
QCAR::Tool::convertPose2GLMatrix(result->getPose());
QCAR::Matrix44F modelViewProjection;
SampleApplicationUtils::scalePoseMatrix(kCubeScaleX, kCubeScaleY, kCubeScaleZ,
&modelViewMatrix.data[0]);
SampleApplicationUtils::multiplyMatrix(&vapp.projectionMatrix.data[0],
&modelViewMatrix.data[0],
&modelViewProjection.data[0]);
glUseProgram(shaderProgramID);
// Draw the cube:
// We must detect if background reflection is active and adjust the culling direction.
// If the reflection is active, this means the pose matrix has been reflected as well,
// therefore standard counter clockwise face culling will result in "inside out" models.
glEnable(GL_CULL_FACE);
glCullFace(GL_BACK);
if(QCAR::Renderer::getInstance().getVideoBackgroundConfig().mReflection == QCAR::VIDEO_BACKGROUND_REFLECTION_ON)
glFrontFace(GL_CW); //Front camera
else
glFrontFace(GL_CCW); //Back camera
glVertexAttribPointer(vertexHandle, 3, GL_FLOAT, GL_FALSE, 0,
(const GLvoid*) &cubeVertices[0]);
glVertexAttribPointer(normalHandle, 3, GL_FLOAT, GL_FALSE, 0,
(const GLvoid*) &cubeNormals[0]);
glVertexAttribPointer(textureCoordHandle, 2, GL_FLOAT, GL_FALSE, 0,
(const GLvoid*) &cubeTexCoords[0]);
glEnableVertexAttribArray(vertexHandle);
glEnableVertexAttribArray(normalHandle);
glEnableVertexAttribArray(textureCoordHandle);
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, [augmentationTexture[0] textureID]);
glUniformMatrix4fv(mvpMatrixHandle, 1, GL_FALSE,
(GLfloat*)&modelViewProjection.data[0] );
glUniform1i(texSampler2DHandle, 0 /*GL_TEXTURE0*/);
glDrawElements(GL_TRIANGLES, NUM_CUBE_INDEX, GL_UNSIGNED_SHORT,
(const GLvoid*) &cubeIndices[0]);
glDisable(GL_CULL_FACE);
// Draw the bowl:
modelViewMatrix = QCAR::Tool::convertPose2GLMatrix(result->getPose());
// Remove the following line to make the bowl stop spinning:
[self animateBowl: &modelViewMatrix];
SampleApplicationUtils::translatePoseMatrix(0.0f, -0.50f*120.0f, 1.35f*120.0f,
&modelViewMatrix.data[0]);
SampleApplicationUtils::rotatePoseMatrix(-90.0f, 1.0f, 0, 0,
&modelViewMatrix.data[0]);
SampleApplicationUtils::scalePoseMatrix(kBowlScaleX, kBowlScaleY, kBowlScaleZ,
&modelViewMatrix.data[0]);
SampleApplicationUtils::multiplyMatrix(&vapp.projectionMatrix.data[0],
&modelViewMatrix.data[0],
&modelViewProjection.data[0]);
glVertexAttribPointer(vertexHandle, 3, GL_FLOAT, GL_FALSE, 0,
(const GLvoid*) &objectVertices[0]);
glVertexAttribPointer(normalHandle, 3, GL_FLOAT, GL_FALSE, 0,
(const GLvoid*) &objectNormals[0]);
glVertexAttribPointer(textureCoordHandle, 2, GL_FLOAT, GL_FALSE, 0,
(const GLvoid*) &objectTexCoords[0]);
glBindTexture(GL_TEXTURE_2D, [augmentationTexture[1] textureID]);
glUniformMatrix4fv(mvpMatrixHandle, 1, GL_FALSE,
(GLfloat*)&modelViewProjection.data[0] );
glDrawElements(GL_TRIANGLES, NUM_OBJECT_INDEX, GL_UNSIGNED_SHORT,
(const GLvoid*) &objectIndices[0]);
SampleApplicationUtils::checkGlError("MultiTargetss renderFrameQCAR");
}
}
// If it was not found exit
if (result==NULL)
{
// Clean up and leave
glDisable(GL_BLEND);
glDisable(GL_DEPTH_TEST);
QCAR::Renderer::getInstance().end();
return;
}
}
glDisable(GL_BLEND);
glDisable(GL_DEPTH_TEST);
glDisableVertexAttribArray(vertexHandle);
glDisableVertexAttribArray(normalHandle);
glDisableVertexAttribArray(textureCoordHandle);
QCAR::Renderer::getInstance().end();
[self presentFramebuffer];
I have tested it in the VuforiaSamples-2.8.9 you have provided as a part of the SDK. Similarly to the vuforiasamples.zip I had sent you the only changes made were the addition of the line you specified and change of the dataset used. Please use the sample application I had sent you earlier. It is not possible that your solution works.
Regards,
W
Thank you for getting back to me, however your solution does not work for me. Within the Session.mm file inside the method of prepareAR() under the [self initTracker]; I have added the setHint. However it still can only work for one object at a time. Any ideas?
Regards,
W
Hi,
After further investigation this does work as expected.
It is however crucial to ensure that the SetHint call to set MAX_SIMULTANEOUS must be done after initialising the tracker.
So in prepareAR() in SampleApplicationSession.mm:
[self initTracker];
QCAR::setHint(QCAR::HINT_MAX_SIMULTANEOUS_IMAGE_TARGETS, 12);
Hopefully, this will work for you, so let me know.
cheers,
N
Hi,
Thank you for the reply. I have already attempted to adjust MAX_SIMULTANEOUS_TARGETS from its inital value of 2 up till 20, all of which have not allowed two products side-by-side to both be 'scanned' within one window. Currently im having to move from one product and focus on another to 'scan' it. I am only using one face/side of the objects side-by-side for this test, with no success.
W
Hi
Yes this should be possible:
Although the link does not mention it specifically you can try increasing MAX_SIMULTANEOUS_TARGETS to 12, though in reality 7 or higher should make this work.
The reason is that each multi-target is made up of 6 image targets, and at any one point in time, only three of them will be in view, and the remainder will be hidden.
Give it a try.
HTH
N
Lol thanks Nalin, will try once I get hold of a machine.
Thanks again,
W
------ UPDATE ------
It worked perfectly!