I haven’t been this excited about something for such a long time!
when I saw that one chain one access have been granted I had to drop everything!
As jen one is hosted on discord, the processes very similar to me journey. You upload a three seconds video (95 frames) and provide either an image, a text prompt, or both.
I haven’t experimented with text prompts very much because I suspect it requires the super long props that work best with Stable Diffusion.
But combining a video with an image is super fun. In fact the process is very similar to the /blend tool in MidJourney, and I can see myself developing a folder full of image textures that are my secret weapons, as some images produce great results, no matter what they are applied to!
I even found a very practical use for a saved GPT3 prompt tool to convert the settings that GEN-1 displays, back into the format which it can accept as a prompt.
The real surprise of this video was the music that was generated via Riffusion, but thats a story in itself…