Sculpture NG from asymmetrical reference images?

[Tseelu]

Hello, I am just starting down the path of creating a 3D representation of Jesus in this painting. I am wanting to do so because ultimately I want to create a “depth map” from that 3D representation of Jesus. Here is an example of a depth map of a horse (Depth map - Horse head by Brallion -- Fur Affinity [dot] net) that I wanted to include, just in case “depth map” goes by a different term. I’m new to 3D. I’ve seen a few videos on blocking for Sculpting. However, most used symmetrical reference images and side views as well. I unfortunately just have the 2D painting I am referring to as a reference image, which has Jesus’s body and head turned along with his arm and hand gestures. I only need to produce a 3D sculpt or model for what is visible on Jesus’ body and clothes. The unseen portions (his backside apparel) is not needed as I will not be animating. I will also not have a need to paint.

I would very much appreciate any suggestion as to what method I should follow for achieving my initial goal (a 3D posed Jesus) that “exactly” mirrors the contours and features of the Jesus in the painting. I’m very willing to follow a video in order to follow the proper way of tackling this project …and then adjust what I am learning through it to meet my needs. I just don’t want to start down the wrong path simply because I didn’t get input from knowledgeable artists like those on this forum. Thank you.

[FedPete]

A “Depth Map” is the outcome of a 3D object.
You can create a “Depth Map” with Blender if you have a 3D model.

Maybe you can create a depth map by hand.

[3DE_Study]

The term is new to me, but seems quite interesting. I did some searching and found that depth maps can be made in photoshop if you have an HEIC file type. I have provided a link to a video that shows this process:

Create Depth Maps in Photoshop | Adobe Creative Cloud

Not 100% sure if this is exactly what you’re looking for, but I hope it helps in someway.

[FedPete]

If I remember correctly, in Blender it’s called something with a ‘Z-…’ .

And there is a physical camera that can generate/use a depth map.
So you can changes focus as a post process.

[NP5]

Well there is the way direct from the image as an ‘alpha’ brush in sculpt mode.
Pretty rubbish but quick and easy, if a little convoluted.

Ideally yes make a full sculpture, It can be non symmetrical, that is only standard as it halves the work to do, you can sculpt with symmetry off, and leave the back as any misshapen blob.

[Tseelu]

Thank you FedPete,
Yes I believe it is called a Zdepth Pass. I have indeed made one in Blender…through following a youtube video. But first I needed to create a model as you mentioned. I’m still at the modelling stage. Thank you for getting back to me so quickly.

[Tseelu]

Thank you so much for the link. Making depth maps in Photoshop is a great choice, albeit it often times is not as precise as what is achieved from trying to make one through the use of a 3D model. If in fact I can’t figure out how to make a proper model of Jesus through suggestions, I will revert back to the photoshop method. thank you.

[Tseelu]

Thank you NP5,
This is interesting (the ‘alpha brush’ option). I will look into it further.

I was wondering your thoughts on my project (and anyone else who would like to chime in). Do you think I should try to sculpt it?

Or, do you think that it would be possible to take a already rigged male 3D model and try to pose it close to Jesus’ pose and then somehow try to adjust (the mesh?) so that it corresponds with the 2D images features?

Or, lastly, I’ve seen a video where someone was not sculpting, but using planes to adjust and build a model (similar in looks to a sculpt)? https://youtu.be/YNgaA4L7Qlo (Timestamp 32:24)

As I mentioned I am new to all and so the path of least resistance, but could adequately satisfy my need, would probably be the path I would choose to take. However, I am very eager to learn what is need to see the project through in its best possible outcome.

Lastly, there is a course by Grant Abbitt that I saw wherein he is building/sculpting a dragon. It’s a premium course. I like his teaching style, but was wondering…if the consensus is that sculpting would be best…and if so do you think that that video would give me the knowledge to get where I need to go? Or is there a free alternative that would suffice. Introduction To 3D Sculpting In Blender - Model A Dragon From Scratch! Thank you for any assistance with any ideas you can share.

[NP5]

Ah well, this is a big question that has no right answer as so often ‘it depends’.
Not only on the quality of the result that is suitable to the end user, but also on the methods that suit the maker’s abilities best.

Now I have a bad tendency of only wanting the very best, this is a failing if time is any issue to you. Mostly far less good is barely distinguishable so it is a failing! So the practical person should try to determine how good a result they need for the use they have for the end result.

There are two basic ways of modelling a 3D sculptural form. Sculpting, which resembles clay working, some even start from a ball, though generally a basic shape of several balls and sausages make a better roughed out starting point.
The other way is as you mention, building the mesh with good topography that aids the underlying shaping as in the video you referenced. It does need good understanding of topology though, how to reduce 3 rows into 1 while retaining quads and such like.

SO much depends on the end use. Most tutorials concentrate on low poly results, as there is a game use obsession, but 3d printing for example wants the opposite, high poly real detail not faked light manipulation trickery. Any route will take a fair bit of time learning. The computer is the tool it will not do it for you by a few clicks.

Yes It may well be possible to take another human model and adjust the model, certainly the body But if clothed that way? May be much harder to find a good starting point one. Changing a likeness, would be hard, not heard of people working that way, probably work for random human variation better than matching a specific one aim, an image like you want. It will always be very difficult to model in 3d to one image. Normally one tries to have front and side on views like technical drawings, and other angles as reference.

That horse head depth map you referenced, It was almost certainly made from a 3d model.
There is a way of taking a picture/render so the closest part of the 3d object to the camera is white and the furthest away is black, therefore depth. The messy quick thing I did above used a black and white version of the Christ image as a faux depth map in effect, trouble is the painting is not shaded on a basis of depth but the artists choice of the darkness of the colours he used.

All of which I am sure is actually no help! It all so much ‘depends’.

[Tseelu]

Okay thank you. I think I will try the sculpting option. I may indeed leave the face featureless. I think because of the image I am printing, the depth related to the face may be undetectable. If I have something recognizable, perhaps I can circle back and ask for additional feedback/suggestions. The journey begins!