I will try to visualise one song with AI, let's see how it goes.
I'll update this thread š§µ
(1/n)
Obstacles 1 Already : Stable Diffusion not starting with custom model.
I'll try for few times if it doesn't work I'll use default model.
(2/n)
Stable Diffusion is not even starting, forget about creating something.
Curr Goal: Get this thing to start.
(3/n)
There's a problem with gradio, other people are also having issue, š„²
From #buildinpublic to #FailinPublic
(4/n)
Atleast I got it to work. šŖš
(5/n)
I can't get a human to fly in sky for some reason.
#buildinpublic
(6/n)
Taking help of #ChatGPT in writing prompts for this song.
(8/n)
Managed to get a style for the whole song.
#buildinpublic #stablediffusion
(9/n).
Getting Good results:
(10/n)
Love is in the air,
Generated 200+ Images today, Experimentation.
#buildinpublic
#stableDifusion
(11/n)
Today I pushed myself to extreme once again.
#buildinpublic #techtwitter #stablediffusion
(12/n)
Tried to animate one of the images, as the song has lots of moments with just beats, static images would look boring.
(13/n)
Day 2:
Extracting All images, will decide on the best ones to use for the music video now.
#buildinpublic
#techtwitter
(14/n)
Seperated Pictures for each song line into different folders, missing pictures for 2-3 lines, I will generate Images for them now.
(15/n)
Using GFPGAN to Upscale The Images.
#techtwitter #ai #stablediffusion
(16/n)
Created New Premiere Project, Imported all the upscaled Images, Now I'll see if it actually works or not. š
#ai #buildinpublic
#techtwitter
(17/n)
Another Failure š„²
I was using Photo Mirage for animation of video but it gives watermark and very low quality output for free version.
I will probably avoid animation for now.
#buildinpublic
(18/n)
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.