How light source size is initialized during training

[pureexe]

Thank you for creating RenderFormer — it’s an amazing piece of work!

I have a question regarding the initialization of light sources during training.

From the paper (Page 5, Section 3.3):

Between 1 to 8 light sources (i.e., triangles with a diffuse emission), with an intensity uniformly sampled between 2,500 and 5,000 w/unit^2,
are placed following a similar procedure as the camera, but with a distance uniformly sampled between 2.1 and 2.7 units.

This describes how the centroid of each triangle light source is positioned, but I could not find details about how the size of each light source is determined. Specifically:

• How is the triangle size chosen?

• How are the three vertices of each triangle light source generated?

I would really appreciate any clarification you can share.
Thank you in advance!