AI applications are only beginning to materialize. It's so exciting what neural networks will do to art workflows in the coming years.
W T F!!! Why, really don't understand, EA = just a fucking looser company!
Amazing... Congratulations for the new way to show information.. I hope they could use this to teaching.
A talented team of 5 artists talked about ‘The Siege‘ project they’ve created for the Unity Neon Challenge. Planning the scene, working on animation, setting up the final look, and more.
For Neon Challenge, we built a team of 5 professionals from the game industry. Jake Ahn is a game designer at Section studios who studied game design at Carnegie Mellon. Han has been working in various industries like commercial, VFX, and now on games. He started as a generalist and now specializing in lighting. He worked on many movie and game titles like Star Trek Into Darkness, Avengers: Age of Ultron, Independence day Resurgence, Guardians of Galaxy Vol.2, Metal Gear Solid V and etc. Companies including Kojima Production, Industrial Light & Magic, Weta Digital and etc. KK acted as the team leader and FX artist for this project. His background is in the VFX industry and since 2016 works for his indie game. Jason Min is an experienced engineer who developed the AKA AR LG smartphone and worked at Zynga after immigrating to the United States 2 years ago. Last but not least, Jason Kim is CG generalist at an advertising company in LA.
There is a music video titled “On your Mark” from the short film 1995 by Hayao Miyazaki. In the intro part of the video, police are using force to suppress a religious cult. We were inspired by this clip and decided to remake it with ‘Siege’ theme.
Initially, our idea was to build a scene with a wide view of a futuristic town and go into a building crushed by a Hovering Police APC (Armored Personnel Carrier) suppressing cultists. However, we only had a month to submit the project so that we started to scope down and focused on the core idea of the scene.
We all are professionals, but this contest was challenging. It pushed us to deliver a core idea using just one scene. Our first brainstorm session was doodling in a cafe. This was all conceived from a messy memo on the back of a receipt written by our waitress’s borrowed pen. Our desire was to create a time freeze scene.
From there, KK and Jake collected references from books and online sources. The “Carousel” and “Cyberpunk 2077” were great ones shared a similar vision within the team.
By our second meeting, we named our project “The Siege” and fleshed out the narrative. Our goal was to answer “What is the cause of the police’s siege?” and “how we can reveal the climax at the end of it?”. After settling on the key concept and agreeing on the shared vision across the team, we started to build the basic settings of our project.
All of us being full-time employees, our time and resources were limited. Aside from the lack of time to create new 3D models and animations, we didn’t have any dedicated modelers or animators. Luckily, we found models of characters and environments that fit our concept from the Unity asset store and Mixamo. As a result, we were able to finish the backdrop of our scene pretty fast.
Our goal was to create a war zone to sell the struggle of the battle. After setting up a first pass of the environment, we filled the space with battle-worn character models.
Then, we created a camera path and adjusted angles to see how the scene would progress.
We used Cinemachine dolly path which Unity3D updated v2.0 in May 2017. Which let us add nodes and optimize the view as we iterated. After creating a satisfying path with the camera, we relocated character models and props that best fit it.
Without dialogue and animation, it was hard to deliver the narrative we created. So we kept positioning character models and props that best conveyed our narrative.
We used the existing animations in Mixamo since we didn’t have an animator on our team. We made a character poses using still frames from those animations. For example, (On the right) the cult leader who is preparing the human sacrifice is originally from a character swing a mace (On the left) from Mixamo as below.
Additionally, the Mecanim’s(Unity’s animation system) animation retarget feature was very helpful.
The very first work we had done in terms of lighting was a post-processing. Usually, post-processing is the last stage on compositing in the film or advertisement industry, but we wanted an accurate preview of the scene from the beginning.Post-processing Stack in Unity3D was pretty convenient to use since almost all the functions were included in one plug-in. Some people tend to use post-processing as much as possible since they think that it just makes the scene prettier, but using too much post processing actually makes the result unnatural. So we use this function subtly and try to make it looks more organic.
Anti-aliasing is used to blur the lines between repeating patterns in order to avoid wavy lines.
Depth Of Field is used to focus on the subject. It blurs out the background from the camera focal point. It helps to separate the subject from the background.
Ambient Occlusion simulates the soft shadows when indirect lighting is cast out onto the scene.
Screen Space Reflection added screen space data to calculate reflections.
Motion Blur is used to simulate the blurring of an image when objects filmed by a camera are moving faster than the camera’s exposure time. This creates realistic motion look in motion.
Bloom is used for creating the glow camera effect on a high light level. The obvious area is the light fixtures on the columns.
Color Grade is used for creating the cinematic look. It helps to change the overall color in post process rather than change the color of lights individually. We made the mix warmer for shadows and cooler for the highlights in this project.
Grain is used to create the real world camera effect that is more common with high ISO films.
Vignette is used to darken image corners when compared to the center.
Chromatic Aberration is characterized by prismatic coloring at the edges of the optical image and color distortion within it. It adds a bit of realism to a real-world camera.
Here is the brief explanation of the lighting setup for this project. As the name of challenge implies “Neon Challenge”, we wanted to create the mood with lighting that is filled with heavy fog and neons. We were deeply inspired by the movie, Blade Runner.
[Blade Runner (1982)]
Our method for the lighting setup is to start with the big lights which will take most of the screen space and then to the smaller ones. In this case, we started with the volumetric light on the main entrance (you can check out the image below). And then working my way out to the doors, pillars, ceiling, and to the main stage (where the ritual is happening). We used the Volumetric Light from Volumetric Lighting Package (Adam’ demo). Because the result of the Volumetric Light was pretty natural, we were quite satisfied with the lighting work.
After we finished the key camera angles and the animation, we started the character lighting. Basically, we used the Cinematic Lighting method for the character lighting and used Local lights (spotlight and point light) to emphasize some parts. We finalized the lighting by adding some on bullet trails and hit positions after completing the VFX.
After camera and lighting were set, we moved on to the effects. The largest focus of the effects was to effectively fill the camera and communicate the narrative of the scene. We carefully set them up to portray the conflict between two groups.
One of biggest challenge was stopping the particle after a certain period of time. This problem was solved by modifying the “Pause” of Particle System.
We like to quote “Small wins lead to big results”. In retrospect, we resolved small problems one by one. These small efforts made our project more polished and fine-tuned. Even earning a delightfully unexpected interview with 80lv.
Also, it was a good hands-on learning experience with the new Unity3D features. Since Unity3D provides a more intuitive editing environment than other game engines, we didn’t spend much time on trial and error. Other may believe in the quote “Think big goals and win big success”. We agree on the first part to have big goals. But to obtain that bigger success, we think you need to scale down by focusing on what is the purest expression of your experience. That is the smallest win first. Whether we hope you all enjoy our work and cheer us on in the near future as we improve our works.