Rookie Awards 2024 - Open for Entries!
My Portfolio
Share  

My Portfolio

by ceadromero on 4 May 2023 for Rookie Awards 2023

I've been interested in character creation for a long time, and over the past year I've been studying and deepening my skills. I wanna show you three characters I've made recently: one for VFX, one for cartoon animation, and one for video games.

6 707 0
Round of applause for our sponsors

Hi, I'm Cesar Augusto Duarte Romero, currently studying at Animum Creativity Advanced School. The following characters that I'm going to show you are projects that I completed, and for each one, I had to deliver weekly assignments until finishing them. Each of the characters came with certain conditions when it came to their creation.

Lizard-man

This project was intended as if it were to be used for a movie, so as a limitation, I had to select a concept art of a creature that could be used in VFX in cinema and had to use a maximum of 8 UDIMS for the textures.
I was looking for creatures to model. From the beginning, I was searching for images on ArtStation that featured reptilian or draconian creatures. Eventually, I came across a concept art by the artist NESSI + that caught my attention because the character looked slim, agile, yet threatening due to its posture. So I decided to start modeling it.

The first thing I did was to sketch some small sketches over the concept art to get an idea, and based on these measurements, I started creating a base mesh in Maya to help me understand the character's form better. Once I had an initial mesh, I decided to move to ZBrush and started giving my character more shape. With anatomical references at hand, I began to mark important muscles and details on the horns, claws, and teeth

After having a more defined body, I started working on the character's armor. I exported the model from ZBrush and in Maya, I began creating a base mesh for the armor piece by piece, with the intention of using it as a base for further sculpting. I made this mesh as clean and defined as possible without adding excessive geometry, and then re-exported it to ZBrush to add more detail to everything.

I performed retopology and did UV mapping, which was done in UDIMs. For this project, I was limited to using a maximum of 8 UDIMs. However, for this character, I only needed to use 4.

I was asked to increase the amount of geometry on this mesh because, as this project was being treated as if it were going to be used for film, the mesh was too low to capture all the detail of the character.

After having my low and high poly meshes approved and completed, I proceeded to do the baking in Substance Painter and start texturing my model. This was a challenge for me because it was the second time I used this program and the detail of the scales was something that concerned me in order to achieve vibrant colors without deviating too much from the concept since one of the main attractions of a lizardman is precisely its scales

Finally, after finishing the textures, I took the geometry to ZBrush to be able to pose it, and then export it to Maya to render it. Here, I used Arnold to be able to output the final render.

For the final render, and considering that it should be a character for film, I used a displacement map to further enhance the scales. In addition to this, the skin has a slight scattering to further highlight the skin and give a better finish.

Rider cat (cartoon)

This is my cartoon project. For this project, I was asked to look for a cartoon concept that could be used for an animated movie. As a condition for this character, I was asked to only use UDIMs for UV mapping and could only use a maximum of 3 UDIMs

I found a concept art on Twitter by the artist David DePasquale that caught my attention. It was a cat with an expression and attitude that stood out to me, and there were many variations and angles that could be useful for the model

First, I took measurements from the concept art to start modeling. I created a base mesh and gradually modified it to get closer to the concept. Once I was satisfied with my base mesh, I moved on to Zbrush and used the Dynamic Subdivision feature to give it a better finish.

At this point, I wasn't sure whether to leave the hair as a volume in the geometry and only work on the texture or to remove that volume and then add the hair later. After giving it some thought, I decided to remove this volume and ultimately generate the hair using XGen.

During the UV creation process, I received a correction to improve the topology of my character's face. I took advantage of this opportunity to also add back the ear hole that my cat had, which I had forgotten to include in the previous step when I removed the hair volume and applied a mirror.

Before texturing, some facial blend shapes were created at the request of the project to test how the mesh would perform.

After that, I moved on to creating the scene, where I first tested the materials and lights before importing the character in T-pose and starting to add Xgen hair. I also created a blend shape to adjust the position of the hair mesh to fit my pose, and finally added some whiskers.

and the final render

GRIFFON (VIDEO GAME CHARACTER)

This model was created to be used in video games or real-time rendering engines, so I was given some limitations when creating it. I could only use a maximum of 100,000 triangles, with no more than 80,000 for the body and clothing, 20,000 for the hair, and 20,000 for props, and work with a maximum of 3 ID maps. For this specific model, I used around 83,000 triangles in total, with 61,000 for the body, 17,000 for the hair, and 4,500 for the props.

For the concept selection, I wasn't given many limitations except for the concept to have good color information and the full body visible. I found this concept art on Art Station by the artist Morry, and I really liked his style, so I wanted to try to translate it into 3D.

In the initial stage of creating this model, I did it a bit differently. I took a human body mesh that I had previously made and modified it in Zbrush to bring it closer to the proportions of the concept. In Zbrush, I also extruded the mesh to create the shirt and pants. Finally, in Maya, I created the props such as gloves, boots, and everything he wears around his waist.

In this delivery, I received a lot of feedback where I was told to pay more attention to the proportions. They also made me understand that the character in the concept art is a bit stylized and its proportions are somewhat exaggerated. Therefore, I had to adjust my base mesh to better match the concept. After making these changes, the difference was very noticeable and I liked it much better. It was much more faithful to the concept.
And at the same time, I made some adjustments to all of his clothing and props to give them a better finish.

And about this model, I added more fine details such as pores on the skin

Then I started to do the retopology of the character, where I only left what is visible on the model in geometry, so I deleted almost everything from the body except for the face. Once I had my topology done, I decided to add small triangle cuts to highlight the wrinkles in the clothing so that it would look much better when backing in Substance.

And then I proceeded to texture the model.

Finally, to finish the character before sending it to Marmoset, I created the hair planes to implement in the model. Here, I used Xgen to create hair strands, using aiStandardHair to give it a good tone, and rendered it. Then, I converted the Xgen into geometry that I used as a high-poly on a plane, and did some backing in Substance where I placed the color render and edited it a bit to generate a normal map and an occlusion map.

I started to add some cuts and placed them on my model here.

And that's it. I posed my character, imported everything into Marmoset, and started arranging the scene to my liking by placing lights. After everything was set up, I created some brick-shaped geometries to give a better finish to the character's render

This is it. These are the three characters I wanted to showcase in this post. I really like the final result of all three, and I learned something new with each one that helped me improve their overall finish. There are still many things I have yet to learn, some of which I may not even realize now, but I want to continue to improve. Currently, I am studying rigging to give these characters even more life.

Finally, I am very grateful that you have read everything, and it doesn't matter if you didn't read it all, the fact that you made it this far means that it caught your attention and you saw the process I followed with these characters, so there is nothing more to say than thank you.


Comments (0)

This project doesn't have any comments yet.