You have probably seen some videos generated by AI and there are definitely some nice ones among them. There is however also a lot of low quality content out there. In this series, I would like to show you how to create videos step by step with AI that tell a story and have a minimum quality standard.
The project
For the purpose of this series, I will try to create a video for the Liotes project. In the context of this project, we initiated a story where a space ship with humans have left earth and are trying to find a planet where they can build up a new settlement. They discovered the planet Liotes. It is a planet with a lot of wind and to start with, they installed power units that transformed the wind into energy. With this energy, they built the initial settlement by melting the sand on the surface and using big 3D printers to build the domes. The last time we have heard anything about the progress of the settlement was in November 2022. Even if it takes long to settle a planet, I believe it's high time that we get news from the settlers on Liotes around Tanja Lemaire.
That is what I want to do through a video that will be generated by AI. The idea is to have a video with several scenes from the settlement, the planet and also with a part where one of the settlers talks to the camera. There will be several scenes that will need to be combined.
With these series, I will take you step by step through the creation process and hopefully we will make a nice looking video.
The tools that I will use
There are plenty of AI tools to generate images and videos and we will need several of them. In order to avoid the recurrent costs of several platforms, I am using a platform that not only allows me to generate images and videos but also gives me access to plenty of different high-end models. The platform I'm using is called Openart. I have a premium plan on it that gives me 24'000 credits every month.
Why not free platforms?
There are quite a lot of platforms that allow you to generate images for free and even a couple of sites where you can generate some free videos. Unfortunately, these platforms cannot compete with the quality of high-end models. For this project, I want to come up with a quality video.
Editing
To edit the video, I will use platforms like capcut or canva that are free to use and allow me to do most of the things that I need.
Sound
The biggest problem I find while producing videos with AI is to come up with quality sounds that match the scenes. For that, I mainly use sources that are free to use but are not AI based like Pixabay. This is not the most satisfactory solution but at least a solution.
How to get started
The first step in the creation process is to know what we want to build. What story do we want to convey and how will we put it into scene.
Even if AI allows us to do a lot of things, it's still a pretty time consuming and costly endavour. Therefore, our video will be rather short probably somewhere between 1 and 2 minutes.
Another point to keep in mind is where and how we will publish it in the end. Will it be a short video or a traditional video? This will define the image ratio which we need to define from the beginning. We will post our video on youtube and on 3speak and we will therefore use a landscape format 16:9.
In the next post, we will get started with building the storyboard for the video.
With @ph1102, I'm running the @liotes project.
Please consider supporting our Witness nodes: