I have a camera device that produce two pictures at the time:
- depth picture with FOV: 87° × 58°
- RGB picture with FOV: 90 × 65°
I expect that they "share" the same objective so their origin can be considered identical. I need to make the images to same size and with corresponding pixels. My idea is:
- Crop the bigger picture to smaller FOV - example for picture width should be $87 / 90 = 0.966$ of the original, so I should crop $(1 - 0.966) / 2 * 100 = 1.7\%$ from every side. Repeat the step for the height of the picture
- Resize one of the picture to match the other (here I lost some information of course, it also depends how I resize it).
Is this approach correct? Do I miss something? Is there a better way to do it?