Revision history [back]

I am streaming data from the kinect for windows v2 into open cv.

I have the color coming in at 960 x 540, but for some reason the depth images are at the strange size of 512 x 424.

I need to process these images by matching pixels one to the other, to triangulate world coordinates.

So, what is the best way to resize the color image to match the depth? This: (psuedo code) cv::Size size (depth.width, depth.height); cv::resize(color, color, size);

Gives an image that doesn't match, due, I think, to the two different lenses on the cameras.

So, how can Iremap the color image to the same size/resolution as the depth, in real time?

Thanks.

I am streaming data from the kinect for windows v2 into open cv.

I have the color coming in at 960 x 540, but for some reason the depth images are at the strange size of 512 x 424.

I need to process these images by matching pixels one to the other, to triangulate world coordinates.coordinates. For this I need the images to be the same size, and the pixels to match up.

So, what is the best way to resize the color image to match the depth? This: (psuedo code) code)

cv::Size size (depth.width, depth.height); cv::resize(color, (depth.width, depth.height); cv::resize(color, color, size);

Gives an image that doesn't match, due, I think, to the two different lenses on the cameras.

So, how can Iremap I remap the color image to the same size/resolution as the depth, in real time?

Thanks.