I wonder if it would be quicker to render them each in the same scene separately then cut a photo of them all into the same scene. Though matching the images with the background to not make it seem so obvious might be a pain.
Then again that may only work with scenes of them just standing separately like in the picture and not interacting physically with each other.
Then again that may only work with scenes of them just standing separately like in the picture and not interacting physically with each other.