Mixing Things Up in Video With AI
Using Luma Ai 'modify video' set on Flex 2 strength with first frames edited in Leonardo (using Flux.1 Kontext).
Here is a fun exercise. Run over to Pexels and find a video clip that you like.
While you can do this with something you record yourself, people often get so self-conscious of their appearance and antics that it freezes them from actually doing the creative exploration. Finding a video on Pexels is a safe alternative.
I found this video of a ballerina by Taryn Elliot of Cape Town, South Africa:
To begin with, I needed the ‘first frame’ from the video. So I opened the video, paused at the very beginning and did a screen capture of the image.
There are a number of ways to edit this first frame. Today I decided to modify the starting image in Leonardo.
I ran the prompt 3x to get the visual I wanted - a more complete chrome variant.
What is nice, is that Flux.1 Kontext only modified the main figure as requested - leaving the background pretty much as it started.
Keep reading with a 7-day free trial
Subscribe to AI Lab to keep reading this post and get 7 days of free access to the full post archives.