Sub·Object

When I began exploring the option of using generative AI to make this movie, I carried over many of the design decisions I had made months ago, when I was working in Unreal Engine: the apartment set layout, the look of the characters, a storyboard for eve…

In my last article, written about six weeks ago, I took a broad look at the gen-AI landscape and toyed with the idea of making the Sub/Object short movie using AI tools.

I'm plugging away at this project during a time when video production is rapidly changing.

Lately I've spent a lot of time in Marvelous Designer simulating clothing.

My current mini-project is a promotional page for the Sub/Object short film.

I thought I might try to sell this laptop asset, so I went through the application process on Fab and learned a few things in the process.

The simplest and most time consuming way to animate this movie would be to shoot reference videos and manually key-frame the whole thing.

Modeling in 3D isn't easy, but it is a simple concept.

Now that I've spent some time immersed in the world of 3D animation, I sometimes catch myself wondering why I never wanted to shoot a live-action movie to begin with.

I moved to Tokyo in 2018, not knowing how long I would live here, and one of my goals was to make a no-budget short film during that time.

After the MetaHuman characters, the two most important assets in this movie are the cluster-computer and the laptop.

Now that I have my shot list, I want to get into animating the characters as quickly as possible.

In my last post about color management, I had decided to export my renders from Unreal Engine in the ACEScg color space.

When I tell people that I am making a movie in Unreal Engine, they often ask about AI.

My current project is to create a sequence for each of the roughly 150 shots in the movie.