Camera mapping workflow

[GIOLETS]

I've think to have understood the correct procedure in order to realize a frontal mapping of a specific image onto the objects of my scene, as shown in the c4d sample scene by Aoktar.

We have 2 choice:

1) create a C4d material and put the image into Color channel, after all we have to do is to choose "camera mapping" onto c4d tag and assign the camera. Using that procedure give me a strange "deformation" error of some zone of the render as shown in my test attached.

2) create an Octane material, and put the image in the diffuse channel, setting "perspective" under texture projection, the camera in the proper field and "world space" into position field.
After that we know to set 180 degrees rotation under UVW transformation and, here is the point I need to clarify, to manually find the correct x or y scale till the image fill completely the screen.

In that way I have however no perfect control on the image size/posizion, and is possible to have significant errorsif between the original image size/position and the camera mapped one.
In all the cases where is needed to make postproduction work and have a perfect matching between original and rendered images, could be a very long procedure to find the correct x,y scale translation values.

I ask myself why is necessary to change the scale of the mapped image if in render settings I have the same output resolution as the original image.
I have missed something? Or there is a rule to apply in order to find these values in a very fast and precisely way?

[aoktar]

You need extra subdivisions to fix deformation errors.