krips891's profile - activity

I am interested in knowing padding and winstride in the function compute. I do not understand their need in getting descriptor for one single particular window. It has nothing to do with the hog literature in general, it is to do with this very particular function of opencv.

Hi can you tell me what are the parameters winstridedoing in the function compute? I asked a separate question by the way: http://answers.opencv.org/question/68...

My usecase is simple. Given an image I want to find hog descriptors of a particular window (that is I just want to compute HOG features in a window). It seems that I should use HOGDescriptor::compute. But I am not understanding what are the different parameters:

virtual void cv::HOGDescriptor::compute (   InputArray  img,
std::vector< float > &  descriptors,
Size    winStride = Size(),
Size    padding = Size(),
const std::vector< Point > &    locations = std::vector< Point >() 
)

The parameter locations makes sense which I assume tells it where to find the descriptors. The descriptors returned would I have a dimension as computed by the the parameters specified by HOGDescriptor constructor:

(_winSize,
Size    _blockSize,
Size    _blockStride,
Size    _cellSize,
int     _nbins ...)

What I don't understand are the parameters winStride, padding. I understand these parameters are used the way a detection window moves for exhaustive detection. But what are they doing in this function compute. Or it is doing something else?! I do not want to compute descriptors in all possible windows, but in a particular window, so that I can feed the descriptor in my custom svm. Thanks in advance.

Great explanation. Thanks a lot. I wish there was a slightest hint towards this direction in the OpenCV documentation. Now the param name win_stride makes a lot of sense, probably one can figure out the meaning from the name itself, but it would still be so nice to have some amount of documentation.

Can someone tell me the parameter win_stride and padding parameters?

I see the parameter block_stride in the constructor. Is it the same as win_stride in the detect* functions? If so, why this duplicate parameter?

I could not find win_stride in the HOG literature. Probably I missed it; it will be nice if someone can explain briefly or point to a source.

I am not able to implement augmented reality using the viz3d module. After getting the pose information somehow (by feature matching) I am doing the following standard steps:

  viz::Viz3d vWindow("Viz3D");  
  viz::Camera vcam(vcammat, Size(640, 480)); //vcammat contains the intrinsic which is calculated beforehand and loaded.
  vWindow.setCamera(vcam);

  vWindow.setBackgroundTexture(frame_vis);  //background frame for augmented reality
  vWindow.setViewerPose(pose);  //pose got by by solvepnp

  vWindow.setCamera(vcam);
  vWindow.showWidget("mesh", vwidget); //some mesh read before
  vWindow.spinOnce(30, true); //this entire thing is in a loop

the pose got is correct as I am able to do a very good overlay of the mesh when I manually draw it in the image using its projected (by pose and intrinsic inside vcammat) points of the mesh. But in the viz3d window I get the window completely occluded by the mesh.

Anyone successfully used opencv viz3d module for augmented reality? Any reason why it is failing ?