BoraxFlux - Sketchbook

This site is terrific. For quite a while I’ve been intrigued by CGI work (and 3D printing). Made the first objects with Blender and followed online tutorials (e.g. Gus -> Blender Wiki). Lurked a lot on this website for the good stuff… the amount of knowledge and quality is humbling and inspiring.

A lizard, made from a cube, subdividing and extrusion and subsequently adding details in Sculpt mode, it has a bone structure for composition:

There are earlier experiments, in which I found that uneven meshes (uneven of density/structure and/or containing triangles or faces with 5 or more vertices) were not good for sculpting. Also when manipulating the mesh in Sculpt mode I noticed when used in extreme, the mesh started to contain gaps and became distorted (I had no knowledge about "Enable Dynamic’ for instance). With the lizard I tried to keep the mesh as even of density as possible, and avoid triangles and faces with 4+ vertices as much as I could.

These are some of those early experiments :

The next project would be printing a Blender object in 3D. I fotographed the head of my hubby from the front, the side and the back and used those photographs as background in blender on all three axes.

I like the details - good work!

You may already know this, but a word of warning regarding sculpting a human head - sculpting in orthographic mode (which is how the background images will be visible, last time I checked) will be quite awkward and will not be conducive to achieving appropriate proportions for a perspective render. I open a number of image panes and load the reference images in there so that they are visible at all times whilst I sculpt in perspective. Alternatively, create a single image in an external application and put that in an image pane.

I had been sculpting in perspective with a focal length of 50mm in the 3D view window (and same focal length for render), but Doris suggested 100mm, which I used this morning and found to be much better.

Thank you for the sound advice Blender Matt! Wish I had signed up earlier for that: for this project I learned it the hard way. My photographs were not orthographic but showed the subject in perspective, which did not work well in orthographic mode indeed. I had to ‘guess’ the average form of the details between front and side view of the photographs in orthographic mode.

Moreover: it’s hard to determine a surface-point of an object in 3D, when the photographs on the x and y axis show mostly soft-toned skin. In hind-sight I wished I had painted reference dots on the face of my hubby before taking photographs (that would have been a lovely job).

In hind-sight I should drop the used method altogether…and use your (Matt) method for similar projects in the future.

Because of my earlier experience with the sculpting mode in Blender I decided to prepare the portrait in such a way, that only little detail had to be added in sculpt mode. I started out with one vertex-point, drew the outlines of front and side- view and filled the model with faces, each and everyone individually by hand… it took frustratingly long. (Extremely grateful for the ‘mirror’ function.).
Building nose, lips and chin:

At one point I even started to browse for information about 3D scanners… I kept making progress however and a big motivation was to see the model 3D printed, the final goal.
The completed portrait in Blender:

A colleague of mine owns a 3D printer (Ultimaker, printed in PLA) and made the physical version, a bust of it is now standing next to my computer at work:

On to the next project =)

Excellent sculpt! Only you can attest to the likeness, but it looks good to me :wink: For a moment there I thought the bottom image was a render, I was going to say that you had managed excellent photo-realism haha. You method sounds taxing, but your results are good!

I am not the best person (by very long way) to comment on all the best ways to sculpt, or the results, I’m still very new to it myself. I’m just passing on information given to me recently. I have learnt a lot in a short period of time reading critiques of other people’s work on here, and then having my own work critiqued (especially from Doris & Michalis), so I know from experience that there is plenty of excellent advice to be found here. FXRs thread concerning the facial topography of Viggo had a lot of useful ideas for me to consider in my own work - I know that saved me from a number of issues that would have consumed a lot of time!

My most recent effort has found me focusing very specifically on the forms that create the light and shadow on the face. I have approximately blocked out shapes on my reference images that show areas where they are lit from above and shadowed below. It seems to have helped me so far to get a handle on the real 3D form from 2D images, perhaps that may help with the issue determining a surface-point of an object in 3D. That may mean you don’t have to paint dots on people’s faces, whether that is good or bad is up to you!