from diffusers import DiffusionPipelinefrom transformers import set_seedfrom PIL import Imageimport torchimport randomimport sslimport osssl._create_default_https_context = ssl._create_unverified_context#locate library#model_id = "./stable-diffusion-v1-5"model_id ="dreamshaper-xl-turbo"pipeline = DiffusionPipeline.from_pretrained( pretrained_model_name_or_path ="../../../../bigFiles/huggingface/dreamshaper-xl-turbo/")pipeline = pipeline.to("mps")# Recommended if your computer has < 64 GB of RAMpipeline.enable_attention_slicing("max")prompt ="R2D2 robots setting a country on fire. R2D2 everwhere. City scape. cartoon. neon bright sunset. 80s. linocut."for s inrange(30):for n in [5,10]: seed = s+21 num_steps = n+1 set_seed(seed) image = pipeline(prompt,height =1920,width =1080,num_images_per_prompt =1,num_inference_steps=num_steps) image_name ="images/synth_{}_{}.jpeg" image_save = image.images[0].save(image_name.format(seed,num_steps))
R2D2 robots setting a country on fire. R2D2 everwhere. City scape. cartoon. neon bright sunset. 80s. linocut. - dreamshaper-xl-turbo
I’ve been having some fun making short animations with Adobe After effects for the tracks on Basement Tapes 1. I started with 7 second animations for the Spotify canvas format. These play in the background when a track is played on Spotify (on mobile I think?). Then more fun was had by lengthening them to 30 seconds for youtube shorts. See this playlist:
I figure I’ll keep making these for some of the tracks on the older albums.
VO: Country Fire
This morning I’m choosing Country Fire from VO. It’s a synthified country/blues song with heavy vocoder. There are periods of rambunctious bleep boops, which sound like a bunch of R2D2s on the prowl.
As a quick aside, I jumped into “Apple music for artists”, and took a quick look at their “promote” features, which produce little graphics like the one in the margin.
Animation details
Feeding this prompt to dreamshaper-xl-turbo:
“R2D2 robots setting a country on fire. R2D2 everwhere. City scape. cartoon. neon bright sunset. 80s. linocut.”
Then, the plan is to mangle these in Adobe after effects…let’s see what happens.
Youtube Short
Very noisy and pixely! I blended through the above images and used the audio to bounce some of the vfx. R2D2s wreching havoc everywhere.