I see a tutorial where they use reference images of a head (photograph) in front and side view. In Blender they go in orthograpic view with the reference images as background and start to model.
I wonder if that doesn’t give a big distortion. Shouldn’t we set the focal length the same as the reference images are? I tried it once with my daughters head and she got then a very tall forehead in Blender.
Orthographic views are a theoretical distortion free views; so the only distortion would come from your reference image. To reduce distortion in your reference image, you’ll want to take a picture of something with a long focal length, such as 100mm or more.
So if you took a picture of your daughter with your cellphone for example, that would create distortion due to the short focal length.
What if I give the view exactly the same “distortion” as the reference image: If the reference image had a 35mm (focal view) and I set my camera in Blender also on 35mm focal view. Should that not solve the issue? I would expect that there wil be a problem when you use photographs as reference (which have ofcourse a lens and therefore a focal view) and you use then orthographic view.
Same distance, same focal lenght, same angle, is what I expect to work, and find it strange I see a tutorial on blender cloud not doing it like that. Or?
You could go through all of this effort, of course, however you will invariably find reference that you won’t know the camera info for. So at a certain point you’ll need to rely less on following the reference exactly, and use it as a guide. This is where your skills as an artist come in, and the more you practice the easier it will be.
So in the end it’s better to work on your understanding of proportion, and scale, instead of fiddling with cameras inside of Blender. I still have a difficult time myself, but it’s really just practice practice practice.
1 Like