How to assign 2d images to planes of a 3d OpenCV Mat, then extract 2D Mats along a different axis?

I want to make a 3D Mat out of stacked 2D Mats and then pull out 2D Mats along a different axis of the 3D Mat. I know I can build this up manually accessing every row or col of every image but is there a better to do it like the following?

vector<cv::Mat> images;

... populate vector with images

int sz[] = {images[0].rows, images[0].cols, images.size()}
cv::Mat cube(3, sz, images[0].type() );
cube = Scalar(0);

for (int i = 0; i < images.size(); i++) {

... place images into 3d mat

开发者_Python百科}

vector<cv::Mat> image_rows;

for (int i = 0; i < images[0].rows(); i++) {

... push image planes into image_rows along width and depth of 3D cube matrix

}

---

I have done a lot of work with opencv and as far as I know that is the best way to do it. You could make a Mat of higher dimensional vectors but then your access would be even less clean. Stick to the vector its your best bet.