Designing a One-Million-Dollar Portrait of Dr. Evil in 3D

Introduction

I am passionate about design, modeling, animation, and the tools that can bring a character and its story to life. From time to time, I like to apply myself and keep up to date with both the graphic software I use and the techniques, pushing myself to learn something new each time, which I can then use in my everyday projects.

About the Dr. Evil Project

My recent works draw inspiration from pop culture, particularly from the characters and scenes of the movies I enjoyed the most: some for the actors or actresses and others for the characters they portray. Dr. Evil, from the Austin Powers series by Mike Myers, has always been one of my favorites. I was particularly interested in capturing his iconic “evil” pose, which embodies a unique blend of malevolence, satisfaction, and humor. To achieve this, I studied reference photos from the films, focusing on a scene that would not only translate well visually but also be instantly recognizable.

Work on the Head & Face

My workflow typically begins with gathering relevant photographic reference materials. These images serve a dual purpose: first, they provide detailed information for sculpting the various facial features, the deformation of facial muscles in a particular expression, lighting information for that scene, and any other useful details; second, I use them for the sculpting session in ZBrush.

I use some of these images for camera matching reference: this is where the real work begins. I started with a base mesh of the head with good topology and skin textures. To find a good camera match in ZBrush, I use at least three reference photos, focusing on the “fixed” points of the face: the lacrimal caruncle of the eyes, the nose, and, if I’m lucky, the corners of the lips, if the subject has neutral expressions in the photos. This is a work in progress in ZBrush before the “evil” expression:

To stay focused on the path ahead, I follow the classic three-step modeling process: primary, secondary, and tertiary. I begin by modeling and sculpting the basic mesh to find the primary shape of the head and face — the larger base forms of the character. During this phase, I further refine the camera matching between the model and the various photos as much as possible. This step is crucial for achieving a believable likeness of the character from all camera angles.

Once I’m close to a match, I shift to a less technical approach, sculpting the character’s unique features with the reference images. Next, I move on to the secondary forms, like wrinkles and other details of the character. For Dr. Evil, the right eye and its corresponding scar are particularly prominent.

For the tertiary details, I use a seamless and repeatable map with skin pores for shading in Maya, which will cover the entire model except for the lips and eyes. Finally, I sculpt the skin pores using Alpha Maps. In Maya, the combination of these two maps will make the shading believable and suitable even for close-up renders.

For each modeling step, I use layers in ZBrush. In this way, I’m able to fix or modify a single layer without messing up other ones. Just as this workflow was recommended to me, I highly recommend it to everyone!

This way not only suggests a clear path to follow during my work but also keeps me focused on what I’m doing, even when I constantly notice other things that are missing or need to be done or improved. Personally, it keeps me focused and free from other tasks that I don’t have the patience to deal with at that moment. Having a clear path for what needs to be done is truly a lifesaver.

It helps me stay focused on what I’m doing so I don’t get distracted by things I see that need fixing — simply because it’s just not their turn yet.

Working Process on the Outfit & Hand

For the hand, I used a base mesh and posed it directly in ZBrush. The outfit was modeled in Blender. I really like this program and want to learn to use it better. So, any excuse is good for a modeling session in Blender.

Nothing special for the outfit: I subdivided a cube to model the basic shapes, set up the UV maps, split them into islands for the torso, arms, and neck, and then exported it to ZBrush. I applied a fabric seamless pattern by aligning the various parts of the outfit, then modeled the shape of the collar to interact with the neck. To give the clothing a believable look, I slightly compressed the neck skin as if it were stressed by the collar.

Finally, I sculpted the seams and fabric folds, positioning everything for Dr. Evil’s iconic pose. The Normal and Displacement Maps were extracted for use in Maya. The ring was modeled in Blender, starting with the primary shapes: a triangle for the top part and a circle for the ring itself.

Texturing

To achieve a good level of realism, I started with scanned maps of the basic mesh. I worked on them in Photoshop, based on the map projection in ZBrush, during the sculpting step. From the reference image, I highlighted three areas of the skin: the top of the skull, which is lighter and more shiny and reflective than the rest, the face in general, and the actual makeup areas of Dr. Evil’s character: ears and nose.

Initially, I emphasized these areas with a slightly bluer tone in the render to give the feeling that they were “slightly rubber,” like in the movie. However, I later decided not to show this in the final render because I wanted a “real” Dr. Evil.

From a Technical Perspective

Once the maps are finished, to make the skin more believable, I usually overlay the Cavity Map onto the roughness, specular, and the SSS map. For skin details, I use two Normal Maps in overlay: the first is extracted from the model in ZBrush; while for the second, I use a seamless map of micro-details that I repeat in the UVs. I use a mask for the second map to avoid affecting undesired areas, like the lips, inside the eyes, and other parts where it simply isn’t needed.

This allows the skin pores, which react less prominently than the rest of the skin, to stand out, thus ensuring a greater level of depth. This effect particularly emphasizes the pores and imperfections of the skin, especially in portrait renders.

During the color correction, when I want to emphasize the skin highlights, using a good clarity value ensures that the highlights are not burned by lights and reflections. Instead, the imperfections and pores of the skin are highlighted in a believable way.

I use the Cavity Map extensively, which I create in ZBrush, especially to define all the details that make the skin look realistic. From certain lighting angles, this technique captures subtle details very well, particularly in the direct reflections of light.

I also sculpted the scar in ZBrush. I kept its color tone slightly lighter than the rest of the skin to give it a subtle prominence. I considered that in the final render, the lighting would still highlight it while keeping the scar in a shadowed area to emphasize it further.

Rendering, Lighting & Post-Production

For the render, I used the V-Ray skin shader, which I really like how it performs out of the box. Combined with custom textures, it provides exceptional skin shading.

For lighting, I used an HDRI with a very low lighting value and a color tone similar to that of the film scene. To achieve the desired lighting, I modified the HDRI by removing the strongest light spots, leaving only a delicate yet believable ambient light that matched the surrounding environment.

After setting the lighting level and HDRI angle, I added a simple lighting setup with area lights. These highlight Dr. Evil’s face, emphasizing the scar, and create a fitting environment for Dr. Evil’s HQ lighting.

To determine the lighting setup, I studied the reflections on the skin and eyes in the reference images to figure out the placement of the film’s scene lights. I had a positive result when I positioned the main light in front and slightly to the side of the face, another light from above to separate the head and chair from the background, and a red light to the right. I made sure to find the right angle so it didn’t overly highlight or illuminate the chair, creating the impression of a well-lit environment.

Tricks for Rendering Realistic Characters

Once I have the final render, I use Photoshop to create a LUT file for use in After Effects. This file contains the color mood I want to give to the image and the character’s turntable video.

In After Effects, I import the rendered layers from Maya: color, albedo, specular, Z-depth, and the one for material color correction: Object ID. After setting up the various render passes in After Effects, I use the Object ID layer for each material or object that needs color adjustment.

I love to use a subtle chromatic aberration effect on my final renders, and I did so in this project as well. From After Effects, I usually export look dev renders like wireframe, full solid, albedo, and others to show some “behind the scenes” processes on social media.

Summary

It took me about 20 days in my free time. My heartfelt advice is to explore. Explore as far as you can, push yourself as far as you can take your skills, patience, and passion to do what you love most. Chart a path, keep your work organized, and reflect on what went great and what could have been better. Every day, you face both theoretical and practical tasks, so acquiring and practicing this knowledge is essential: in character creation, anatomy is the golden key. It is important to be happy and enjoy the good things you achieve, then work on the ones you need to improve.

Sneak Peek

Another portrait I’m working on these days, same process, same apps, still in progress:

For more behind-the-scenes looks and updates on my latest projects, follow me on Instagram. Thanks a lot, 80 Level!

Emiliano Calisti, Graphic Artist

Interview conducted by Theodore McKenzie