Hello everyone.
I been practicing to make an anime face, but i just cant seem to figure out how to make my anime face nicer.
I know something is wrong, but i just dont know what.
I took reference from the following video.
Can anyone have a look and help me out?
Thank you.
(With Subsurf modifier)
(With no Subsuft modifier)
(front)
ok ill try to show some of what i thin,
1- ther is not enough vertes, to make the fase smooth, or to even creat a face
2- subsurf makes thing even smoother than it should, so you need to add more verts to control it,
3-the area wher the ears starts, to the top of the head sholdnt bee that smooth, its almost sharp ,
its beast to model without subsurf thin add it later,
and also its beast to model from am map or an img refrence
@Amiyul_SparemeAlife can you post a shot of the wireframe in edit mode? Looks like the loops around the mouth up the cheekbones and under the eyes is a bit too flat and pinched. Hard to see with the topology if you just have the loops too close or are pulled too far out at the edge causing that sharp transition.
This has a good example of a simplistic topology that works. https://www.foro3d.com/f215/character-design-2d-sketch-to-3d-76478.html
Topology is a important for animation, UVs, and I fond it necessary for primitive modeling. Also knowledge of anatomy can help you with stylized characters that don’t have defined features.
http://thundercloud-studio.com/tutorial/modeling-guide-realistic-human-head/
There’s a gentleman on Pinterest site who has quite a bit of a topology collection. Some images might be of interest to you.
https://www.pinterest.com/kelleywilliams3/3d-modeling-head-face/
Welp, there’s a lot of room for improvement, if you ask me. Like e.g. the nose from the image above pointing to this post. It’s point- ing. Meaning how could you even change the shape of one vertex, but, well.
Have you tried modeling combined with sculpting in Blender yet? I’d suggest give sculpt a try; make a copy of your mesh on another layer, switch on Dyntopo and see if you can pull the shapes you are looking for. After you can either delete, re-skin to lower poly or vertex paint and use sculpt for renders.
Sorry for the delay guys, had some stuff to settle at home.
Alright! Thanks so much for helping me out guys.
I did improvements on my old face.
This is what I have, feel free to be as honest as possible cause
I really want to learn more on how to make my face as pretty as it can be,
before I start doing my favourite part, which is animation. =D
(Front)
(With no Subsuft modifier)
(Front)
(With no Subsuft modifier) (Colour)
(Front)
(With Subsuft modifier) (Colour)
(Side)
(With no Subsuft modifier) (Colour)
(Side)
(With Subsuft modifier) (Colour)
(Top)
(With no Subsuft modifier) (Colour)
(Top)
(With Subsuft modifier) (Colour)
(Wireframe)
(Front)
(Side)
(Top)
hey man, looks like you’re just getting started modeling, and you’re doing a good job so far! Especially with faces, to get good with this stuff is a lot of work. When I was getting started, I modeled around 5 crap faces for 1 that I was satisfied with. This is already a lot better than some of my first models, and you’ve also done a great job adding in the topology!
So the first thing is looking at the overall shape of the face. Here, having a end goal really helps. You should find reference of what you’re trying to do and try to emulate it. Here, I just overlayed a picture I found on google.
This is a picture I found, but you should do this with a picture you’re trying to emulate. When you’re getting started, it’s good to use other people’s art, so you start to learn shape language and stuff from professionals who have been doing this for years. Having reference is how we learn as artists.
Immediately, though, there are some things we can see that would make your model stand out more. Bringing the nose down, and pushing out the curve of the forehead would help. The neck is really far back, bringing that forward would help balance the head overall. These are things we might not notice without having reference.
A tool I like to use to help me visualize is liquidify in photoshop. I think this is called IWarp in Gimp? it lets you push around bits of an image, so you can start to shape things. Here I did it with your model, to try to get it closer to the reference.
And here we can immediately see the second problem. the resolution of the mesh is really low to get the shapes of the face. You can see there is one face between the end of the nose and the lip, which is not enough resolution to capture the shape of the lips and the nose. I think you could easily double your face count (quadruple even in the mouth area). As mentioned above, the subsurf modifier is awesome, but tends to goop things out, instead of make them super awesome. 
hopefully that helps! Good luck!
This is a good start but I recommend poly by poly modeling when working on the face. box modeling is good for getting a overall shape of the character and base meshes but it wont help when you need to work on topology. You will also need more verts to get a more accurate depiction of a head.
like pancake said, you should definitely be using a model reference at your skill level. If you feel like it is cheating, it’s not looking at a reference while trying to model will help you get accurate shape and will be easier to tell if something is off. You can work off of memory for a challenge when you are more developed. Using the reference will also help train your 2D shape to 3D form translation skills. This video will give you a good idea of 2d to 3d translation https://www.youtube.com/watch?v=WoyQGlVgF0o
These are two most common was to modeling with reference images in blender.
Setups.rar (315.5 KB)
Wow, thank you soo much guys (Pancake and Dan), for the feedback.
It really helps.
I’m happy that there are supportive and helpful people in this community. =D
I’ll be sure to make full use of all the advises and tips given in this post.
It wont be easy, but I know that I can produce amazing work too, as long as
Im working hard and have patience. 
Lets goo!!
Then again, topic says “Anime face”…
Ain’t we pushing things too far into some portrait topology which would need a ton of well defined and calculated amount of properly placed loops?
What if to follow KISS (principle, if what) and limit to just what’s needed to animate eyes, mouth? Like this e.g. which is pretty close to what OP has now. Shape, proportions - either sculpt or proportional edit, done.
