0
$\begingroup$

so I am trying to model a bust of Jon Bernthal's Punisher and I couldn't find a side and front reference of his head in the same picture so I had to get 2 pictures from the same photo shoot but I am having a hard time getting both reference views the same size for modeling. Is there a trick to get them equal size or way to model the head with only one view reference? Any help would be greatly appreciated since I'm still kind of new to Blender modeling software.

Jon Bernthal front view Jon Bernthal Side View

$\endgroup$
3
  • $\begingroup$ Hello and welcome, your question is basically not about Blender. But a where to get an asset. Things that you could is to photoshop the images and try to to get the same size for the head as much as possible. $\endgroup$
    – Emir
    Commented Aug 11, 2023 at 17:12
  • $\begingroup$ align the eyes with the eyes, the chin with chin? $\endgroup$
    – moonboots
    Commented Aug 11, 2023 at 18:03
  • 1
    $\begingroup$ If it's the same photoshoot in similar position then they should be the same scale, no? To scale references the easiest you can do is put one on top of the other (With one's opacity lowered) and then align some common reference point on the outline of the head, like the chin. Then set the Pivot point to Cursor, put the cursor on the chin and scale or rotate the side view until other reference points like the ears , hair line or eyes are at the same height. This is an approximation tho, photos distort the image depending on the FoV so it's never 100% correct $\endgroup$
    – Cornivius
    Commented Aug 11, 2023 at 19:02

1 Answer 1

0
$\begingroup$

I would set up two cameras with the reference images as background images and a human dummy model. So it's easier to guess the point of view, focal length, and so on.

camera setup with background images

The first image (front view) is a half-body portrait. The focus is on the chest and you can see there is some perspective/distortion. The head seems to be lifted up a bit because of the perspective. This is influenced by the camera's and viewport's focal length.

Since you don't know the used real camera's focal length you have to guess here a bit. Checkout common values for portrait shoots (50, 85, 105, 120, etc).

For sure the head is tilted a bit to the right, so roll the camera a bit.

You can play with the camera's background image settings (front/back, opacity) and the viewport's overlay (wireframe, x-ray) to get an idea if the model/camera position matches the reference image. Set the Frame Method to Fit so the image is not stretched.

Enable Lock [X] Camera to View in the side panel to move the camera. Use the Transform > Rotation values for the camera to mimic the head tilt.

front view match

Make sure the eyes, the center of the mouth, the nose and the ears match the reference as closely as possible.

In the side view, the head is not fully perpendicular to the camera. Look at the philtrum (the area between the mouth and nose) and the eyebrows. Move the camera and try to match the inner ear (auditory canal), the eyes and mouth of the reference to get the correct point of view. Play with focal length and the distance between the camera and model (zooming).

side view

Once you are happy, disable the Lock [_] Camera to View in the side panel so the camera does not move when you pan around in the viewport. Now, you can start modeling and sculpting. Look through the cameras to get the front and side view (Numpad 0).

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .