logo80lv
Articlesclick_arrow
Talentsclick_arrow
Events
Workshops
Aboutclick_arrow
profile_login
Log in
0
Save
Copy Link
Share

Creating Kung Fu Panda's Tai Lung in 3D with ZBrush

Sarah Nalaar spoke with us about recreating Tai Lung from Kung Fu Panda with ZBrush and Marmoset Toolbag, transforming the character's 2D concept art into a fully functional 3D model.

Introduction

Hi! I'm Sarah, a 3D Artist from Germany specialising in stylized and low-poly game graphics, with a strong focus on characters and creatures.

I came to 3D through experimentation in Blender and subsequently began studying to become a game artist. Over the past six years, I have started developing advertising for XR apps, been involved in indie games, and discovered my passion for character art.

Getting Started

Alongside my freelance work and previous full-time jobs, I regularly work on personal projects.

For Tai Lung, I drew inspiration from several pieces of art that looked like 2D illustrations at first glance but turned out to be created entirely in 3D. I wanted to achieve the same illusion, especially because I love hand-painted textures and use them regularly in my own projects. 

I had wanted to adapt a concept by Nicola Saviori for a long time. His designs and his painterly style seemed perfect for a completely hand-painted project, and I am a big fan of his work, too. 

The choice of Tai Lung was very personal. As a child, I loved Kung Fu Panda, and Tai Lung has always been one of my favourite villains. Coincidentally, he is also Nicola's favourite film villain, which made the choice even more fitting.

Modeling

I began the modelling process with a very rough blockout, focusing exclusively on the main shapes and overall proportions. At this stage, everything was deliberately kept loose and simple.

One of the first challenges was to translate the strong perspective of the concept into a 3D model. My original idea was to achieve this effect later through the camera's field of view, so I kept the proportions closer rather than exaggerating limbs or body parts too much in the sculpture. Over time, I realised that this approach didn't work as well as expected, but it allowed me to create a solid foundation before I started working on the shapes.

Spoiler: Only at the very end, when I was already in Marmoset at the render stage, did I realise that the FOV perspective wasn't what I had imagined, which is why I adjusted the sculpt with texture again... but I'll come back to that later.

I did not retopologize because the model was only created for texturing practice and rendering and was not intended as a game-ready model.

Texturing

The entire texturing process was made directly in ZBrush, using a hybrid approach between projection painting and hand painting.

First, I assigned flat base colours to each sub-tool in order to clearly separate materials and shapes. Then I imported the concept art as a texture and added it to the spotlight. From there, I was able to move the Concept around and paint important colour information directly onto the Model with the paint brush. To achieve a clean and legible result, I worked with the "Flat Colour" material and the standard brush, blocking out base areas from which I could later select colours.

Since the projection quickly becomes distorted due to perspective and surface curvature, most of the work was done by hand-painting. Viewed from the front, this was relatively easy, as I was able to stick closely to the concept. However, the rear view was much more difficult, as I had to invent colour transitions and details while maintaining the style and lighting logic.

I repeated this process for the entire model until the textures appeared uniform. For the single fur strands, I added individual strands of hair as separate sub-tools to have more control over the shape.

Lighting & Rendering

The final image was rendered in Marmoset Toolbag 5 using ray tracing. The materials were kept very simple: colours were applied using vertex colours, and the material itself was set to "unlit" to maintain the hand-painted look without realistic shading.

For the lighting, I used several spotlights, which were mainly used as edge lights in order to stay as close as possible to the original concept. I deliberately chose not to include lighting information in the textures, as I wanted the model to work well and be easily readable from all angles.

The outlines were created by duplicating the model, mirroring the surfaces, and applying a dark blue material. I selectively scaled the outline mesh, extending it only in areas where outlines were needed, while leaving other parts unchanged or slightly reduced in size to avoid outlines there.

First, I experimented a lot with the camera's field of view to recreate the perspective of the concept as I had originally planned. When that didn't work visually, I reset the field of view to the default setting and went back to modelling to adjust the proportions directly. The result is that the model looks correct when viewed from the front, even though the rear view now has slightly exaggerated elements, such as Tai Lung's oversized foot.

In post-production, I added a subtle film grain and a light depth of field to finish the image and give it a more cinematic feel.

Conclusion

One of the biggest challenges was translating the concept's extreme 2D perspective into a functional 3D model. I initially tried to solve this through camera angles, but found that such a perspective needed to be built directly into the sculpture by using exaggerated proportions rather than relying solely on the field of view.

Another challenge was texturing outside the areas shown in the concept. While the front view was clearly defined, I had to adjust colour gradients, brush strokes, and details for the back of the model while staying true to the original style.

Throughout the project, I learned a lot through trial and error. Even when something didn't work, I was always able to take something valuable away from it.

My advice would be: Do not be afraid of making mistakes and feel free to repeat yourself – every mistake is an opportunity to learn something.

This project was both a creative challenge and a valuable learning experience that emphasised the importance of experimentation and, of course, a lot of fun!

Sarah Nalaar, 3D Character Artist

Interview conducted by Gloria Levine

Ready to grow your game’s revenue?
Talk to us

Comments

0

arrow
Type your comment here
Leave Comment
Ready to grow your game’s revenue?
Talk to us

We need your consent

We use cookies on this website to make your browsing experience better. By using the site you agree to our use of cookies.Learn more