Question about rotation and translation data format in OmniBlender/Ricoh360/OmniPhotos/360Roam/OmniScenes/360VO datasets

[shiming2014]

First, thank you for providing these valuable datasets! I'm working with the datasets mentioned in README and would like to clarify the coordinate system definitions in the camera pose data.

Could you please confirm:

​Rotation matrix (R): Is it defined as a ​World-to-Camera (W2C)​ transformation (i.e., transforms points from world coordinates to camera coordinates) or ​Camera-to-World (C2W)?
​Translation vector (T): Is this the camera position expressed in ​world coordinates, or is it the translation component in camera coordinates?
This information is crucial for proper alignment in our 3D reconstruction experiments, and we want to ensure we're interpreting it correctly.

A minimal example or mathematical formulation would be especially helpful. For example:

If W2C: X_camera = R * X_world + T
If C2W: X_world = R * X_camera + T

Thank you for your time and clarification!