0

I have a CAD model of an object in Blender software. And I had printed it out and took some photos of it.

I need to use the Blender software to render some synthetic images of the object. My task is to create rendered images as similar as possible to the photos which I took for the printed model. In the Blender software, there is a camera view to capture images to render. What I have are:

  1. CAD model of the object enter image description here

  2. photos of the object enter image description here

  3. background photo enter image description here

What I get now is: enter image description here

But obviously, the perspective and orientation of the 3D model in the synthetic image are not as same as the real photo's. How can I do to match them?

wangmyde
  • 77
  • 8

1 Answers1

0

Normally, this is a problem of different interpretation of the camera coordinate systems. For example, opencv interprets the z-axis or the camera to be aligned with the optical axis of the camera and going from outside to inside. You can check how the blender interprets the camera coordinate system and transform it to match opencv.

Derza Arsad
  • 41
  • 1
  • 2