Creating a quality video with AI – Creating image material

avatar

$1

In this post we will work on the 3rd scene of our project. In our storyboard, we had the following content of the 3rd scene :

While Tania is still talking we will take a shot of the settlement to illustrate what she is telling.

To define the content of this scene, we need to know what Tania is speaking about. Which is the size of the settlement and the domes. According to Tania, there are now 18 Domes in the settlement. So I would like to have an initial image with 18 Domes that I can later animate. However, we have already some image material of the settlement that I want to reuse.

We have this image here that was illustrating the settlement Leonopolis so far:

$1

Now in Openart, I use the function Chat to Edit with the Flux Kontext Model. I simply feed the image above and ask AI to add more Domes so that we have a new image of the settlement with a total of 18 Domes.

I needed a couple of tries until I got the image that was close to what I expected.

$1

The text of Tania for this scene has a duration of 18 seconds. So I will need a video of the same length or several videos that together make up at least 18 seconds.

In the first shot, I will use this image and animate it with the Pixverse model with the following prompt:

The Camera orbits right around the settlement.

This gives me the first 5 seconds of the video. To extend this video, I will then take the last frame of the video using the website https://finalframe.ai/frameextract/. This will then be my first frame from a similar shot where we zoom in on the settlement to give the viewer the feeling that we go closer.

Tania also talks about the different use cases of these domes, so we want to produce some images from the interior and then animate them.

I will create the images from scratch with text to image and I will use Flux pro with the following prompt:

Inside view of an agricultural unit in a dome on a different planet. The dome is about 30 meters in diameter. There are some workers in white and orange combinations. The dome walls are milky white semi transparent.

It needed to generate about 20 images until I found something that I liked. I did the same process for the residential domes and also for the common dome where people can eat and spend their free time.

Here the different images that I have created and that I will later animate and put together.
The agricultural Dome:

$1

The residential Dome:

$1

The common Dome:

$1

Now that I have these images, I will animate them so that later I can add them together to build this scene completely. I will use the same process for the scene 4 as well.

In the next post, we will look at scene 6 where Tania speaks about the Ark 62 and the Jumper vessel that goes from the surface of the planet to the Ark.

Check the previous posts of this series:


With @ph1102, I'm running the @liotes project.

Please consider supporting our Witness nodes:



0
0
0.000
9 comments
avatar

With the help of this software, we can now create many things easily.

0
0
0.000
avatar

It's nice to see the progress. AI is working out well to give you options, and its slowly building up to the final results.

0
0
0.000
avatar

It's nice to know all the tools that allow to do this step by step process.

0
0
0.000
avatar

Hello!
Although I don't have the opportunity to learn about these topics yet (due to technological limitations), I always enjoy these types of articles that introduce me to new learning.
Greetings!

0
0
0.000
avatar

This looks pretty good actually. You might even be able to use this as a pitch to gamify Liotes somehow! 🤔

0
0
0.000
avatar

I have to admit that I needed to generate a lot of images to find the ones that I chose here but it's true that with more images, the whole concept of Liotes takes on new life and it could definitely be used for further gamification ;-)

0
0
0.000