Lazar Damjanov is a character artist at 3Lateral studio, a company known for creating digital human characters mainly in the game industry. He has a bachelor degree in Graphic Engineering and Design at Faculty of Technical Sciences in Novi Sad, but as a result of his interests in 3D modeling he joined 3Lateral studio in 2012 and has been working at 3Lateral for over 6 years now. During that time he has been involved in many AAA projects mainly as a supervisor facial shapes modeling artist. Those projects include games like Insomniac’s Spiderman, Horizon Zero Dawn, Star Wars Battlefront II, Red Dead Redemption 2, Battlefield 1, Batman Arkham VR, Until Dawn, Star Citizen, Hellblade, The order 1886, Ryse: Son of Rome and many more. His interests also include character concept art, character modeling and texturing and long walks on the beach 🙂
Welcome to CGA2018! Do you have your slippers on? Well, you’ll need them as we are going for a stroll.
Hey! It’s nice of you for having me and if you don’t mind I brought my own Batman slippers with me. I just feel more comfortable and confident wearing them 🙂 Also, I will be doing the whole interview in my Batman voice if you don’t mind.
Facial modeling artist sounds quite exotic. What is it all about?
Some people say that if you don’t know how to explain something the easy way, it means that even you don’t understand it quite well. After years of being in this industry and explaining my job to others, people still think that I am a programmer. Jokes aside, facial modeling is all about making human expressions and all facial deformations believable. This is one of the few positions in our pipeline with the tag “artistic”. Every character that goes through our pipeline (digital double or fictional) has some sort of artistic interpretation embedded in it. My job is to make sure that every possible facial expression looks good. There it is! The simplest explanation! 🙂 Sculpting expressions is more difficult than regular static sculpting, because you have to make sure that transformation between neutral pose and the expression itself looks good. These expressions in our rig are not treated like emotions, but as a mechanical deformations defined by Dr. Eckman in his FACS system. FACS is an acronym that stands for Facial Action Coding System and it is foundation of every digital character made in 3Lateral. Mixing those FACS expressions you get emotions. Often, I feel like Dr Frankenstein, bringing neutral heads to life! As I mentioned earlier, there are two types of characters: digital double, which is the accurate 3D replication of some person from real life, and fictional. Every type has its own pros and cons. When I am working with the scanned actor, I get certain number of scanned expressions (the most important ones) and I build other expressions around them in order to avoid the uncanny effect. Fictional characters are more fun to work with, because you build your own expressions using your experience in order to get likeable character.
How did you get interested in a specialized and exciting field like facial shapes and how did you progress to work on such high-level game titles?
By coincidence! Simple as that. I came in 3Lateral in 2012. with some of my drawing when the studio was searching for digital artists not knowing any of the software needed for the job. At the interview, I even hit my head on the doorpost. Pretty sad. Today, I like to joke about it, that I got the job only because of that. I was so frightened that when I got home I didn’t even know how the studio was called. This is a true story! Early days in 3Lateral were all about training and getting used to working with Wacom tablet. In that time, the need for the facial blend shape artist was starting to build up, so naturally, I pivoted my training to learning Maya and facial anatomy.
I was lucky enough to work for some of the AAA titles right away and learn through them. My first ever project was GTA V where I sculpted wrinkle maps for the in-game characters. After that I was just trying to learn the fastest way often not knowing what projects I was working on. When you are given this kind of chance and you get to learn working on Gary Oldman or Rami Malek, you have to get better at it. I think these AAA titles that came along the way are just some sort of evidence, that we worked hard as a team.
What would be divide between creative and technical aspect of your daily work? What are the skills needed for both sides?
There is no division between these two as they are intertwined. Artistic aspect is not existent without technical knowledge. Only after few years in 3Lateral I could enjoy artistic side of my job as I was struggling overcoming the software and tools in it. Technical knowledge is just a consequence of pushing your artistic limits further. I think that artistic in our case means advanced perception of human face which is constantly evolving day by day, project by project. Our artists and me personally use all sorts of software like ZBrush, Maya, Photoshop, 3DS Max, Unreal Engine plus many more. References are gathered most commonly as strange tabs in our browsers such as “baby smile”, “girl arms” and “horse muscles”. The most important thing that every artist must know is the technical side of our pipeline. Getting the bigger picture and always questioning your work methods. And having fun doing it!
What was your main research to learn about the human face and how complicated is to know it in perfect detail?
At first, I was learning from our CEO, Vladimir. Our studio was smaller at the time, so I had the privilege to learn from him through our daily sessions. That was the time I first heard of FACS systematization of human expressions. The most valuable lesson that I remember from that time is that you really don’t have to know all those Latin names of facial muscles, but only their position and try to predict what will it’s contraction do to the skin above it through the process of sculpting. Later, I researched the emotional side of human face through another of Dr Ekman’s books titled Emotions Revealed and currently I am using Anatomy of Facial Expression to get a better understanding of muscle structures. The thing that is the most interesting is that I am constantly researching human face when talking to people. For instance, a few weeks earlier I noticed that the tip of my colleague’s nose is going downwards when he is puckering or pushing his lips towards generally. That sort of reference is the key to making digital people more convincing. Of course, currently, I am limited as I work only with blend shapes which is a simple morph between neutral head state and expression. Every vertex moves linearly, so I have to incorporate every single movement that i find interesting in order to convince people that the character have just winked at them 🙂
There is so much more to know about the human face. Personally, I really like mixing certain expressions and making strong emotion from them. Especially if that character is fictional. There is something more satisfying in that case because not only he doesn’t really exist, but you’ve managed to derive emotions from it and make him funny or evil, or both! As our pipeline constantly progresses, the understanding of facial deformations will be even greater. I can’t wait to learn more about micro expressions and import them in our characters! The thing is that every person out there can be the judge of our work and our job is to thin the border between “naaaah” and “wow!”
Do you record your own references, or is it too weird to stare at yourself?
Every modeler at our studio has a mirror on his desk. It is perfectly normal to see someone looking in the mirror opening his mouth like a fish on dry land or just frowning at himself or herself. Also, I have pictures and recordings of members of my team as part of the exploring certain features on their faces that just look cool. One time I tried to see how my blink looks like in the mirror and forgot that I will not be able to see anything since my eyes were closed… Ah, that struggle 🙂
There are not only human faces in games, what about alien species? How do you find references for them? Any secrets that the CIA should know?
That depends on what alien species we are working on. You see, there is a top-secret work and travel program for students financed by 3Lateral with the main goal of sending students on the planets in galaxies far away and acquiring precious knowledge about those species and sending reference videos to our team here back on Earth. I am just kidding and at the same time demonstrating one thing you can turn yourself to. And that is imagination! Most of the alien species we were working on were humanoid characters which mean that same rules apply more or less. You have to adapt your work based on your experience and what you are working with. Some of the most intriguing alien characters that I’ve worked on are Maz from the Star Wars franchise where I had a bunch of references from the movie and our Osiris Black alien seen in our tech demo with Andy Serkis. The second one was more difficult for sculpting expressions as I was constantly discussing possible solutions with our art director who made the thing. And CIA, well they are also familiar with micro expressions during interrogation process, so they are good 🙂 Maybe I should tell them that we are not alone in the universe, but I think they know it already.
Working as a specialist means working in a team and teamwork. Where does facial artist sit in the production pipeline and who are the main collaborators when working on the project?
Facial modeling team consists of people with many interests and everybody has its own take on the job. The team is constantly being changed due to opportunities that are given to people based on their interest. And me, as a supervisor I was working on a personal cartoony character for over two months this year, learning new stuff and applying earlier gained knowledge here in 3Lateral.
Facial modeling comes right after the scanned data is being processed and neutral head scene with all assets is being generated. It is very important to have an open communication with the scanning team in order to achieve greater likeness in every single acquired expression. Often, the tiniest misinterpretation can lead to strange results afterwards. My task is to make sure that every single data from talent’s face is being imported in our rigs. As accurate as the scans are I can get a better sense of characters expressions later in modeling process where I have to imagine what his face would look like in some of the intense emotions and combinations of few basic expressions. After my job is completed, the rigging department is taking the sculpts and generates a final rig. Here, in the same spot, between scanning and rigging.the majority of art department sits. Later, if needed the rig is being polished by our look development artists. So we are basically living organism with different layers of artists, technicians, riggers and again artists and all of this is supported by programming team.
How long does it take to create the digital human and what is the established process?
Hm, that’s a tough one to answer. There are many parameters that affect the speed of generating a digital human. For example, there is a huge difference between some random NPC character and the main one. But, when you have an established pipeline we are talking about weeks, sometimes even days! Our last tech demo with Andy Serkis was done in just few weeks (only Andy’s face), so sometimes there are no rules. If we are talking about game project, usually the master character (the first one) was done in time needed to achieve a desired look and that is time-consuming process. Later on, when the pipeline is established, everything goes much smoother and faster. An established pipeline is the one with calculated times for every member of the team. It is very important doing your part of the process on time, as we are working with the files given by previous team member. I’m pretty sure that even Terminators will have our rigs on their faces if we continue making this rapid progress 🙂
Fast-paced technological progress together with 3Lateral on the frontline is pushing boundaries of digital human creation. How do you keep your skills up up-to-date?
Yes, the boundaries are constantly being pushed further away technologically speaking, but we are lucky that evolution is a pretty slow process which means human face is not being changed drastically 🙂 We are just finding new ways understanding and describing it. As the new tools are being developed, that merely means that something big has changed in a bad way, it is only making our job easier by adding up on the previous knowledge. It is just like a wax on-wax off scene in Karate Kid. Few basic moves and off you go! Fight!
Working on many AAA games characters, how do you see your characters when you actually play the game? Can you distance yourself from your work and see characters as a part of the story?
Well, I am not a big gamer, but I made a promise to myself to buy every single game I was working on. And no, absolutely not! I cannot distance myself from my work. Even when the trailer comes out it is a big day in the studio. Later, when I am playing the game I am usually looking for any possible mistakes in character’s performance. My own job is preventing me to fully enjoy in these truly magnificent titles that came out through the past years, but hey, what are you gonna do? We are making a world a better place 🙂
Every CG artist has his precious personal projects, can you share it with readers?
Earlier I mentioned cartoony character I was working on this year. My personal preference are stylized characters because I think that in their simplicity is hidden complexity of well-thought proportions, different shapes, color patterns, movements… This character is called Ćoravi Žika and it started as a personal project which I thought it would be funny to do facial shapes for. I wanted to explore how far I can push my shapes modeling skills and to test myself in that way. One good thing that happened through this process is that I pulled two more people with me. One person was working on hair model and another one was working look development in Unreal Engine. So, this personal project has become studio project as many more people got involved. In the future, I hope that I will do some work of this kind as I want to explore facial expressions on stylized characters. Also, my personal work includes drawing (mainly digital) and sculpting fictional characters and my interpretation of existing ones.
What is your drug of choice, besides staring at human faces? How do you spend time outside of the digital world?
I fight crime in the streets when everybody is asleep! The first thing that you are going to hear about me is that I really like to ride my bike. If you ever need the guide for Fruška Gora and its surroundings, I am the guy! 🙂 My interests also include tv shows, going to movies, concerts, spending time with my family and my dogs and trying hard to be the funny guy (and yes, I am aware that the last sentence is pulled right out of the dating sites).
What are your expectations for this year’s CGA2018?
This amount of knowledge in one place can’t produce something bad! I really wish that our community will become even bigger after the conference as many young enthusiasts will hear some of the best people in the trade.
It was great talking to you and it was a nice stroll too. Thank you and see you at CGA. Now you can stop staring at my face.
Oh no! I lost my Batman slippers! Professional deformation, losing slippers and staring at the people in the streets or in the bus…or on the bike, but that is a story to itself.