Used SwarmUI’s Segment Refinement to get the image on the screen, then added a small amount of static with GIMP.
Prompt: “a television < <p> <segment:screen,0.8,0.4> a nude woman dancing, no text on the screen”,
Model: “stoiqoNewrealityFLUXSD35_f1DAlphaTwo”, Seed: 214070023, Steps: 15, CFGscale: 1.0 Width: 1152, Height: 896, Sampler: “euler”, Scheduler: “beta”, Flux Guidance Scale: 3.5
This is a neat idea. I’ll have to look at segmentation nodes for Comfy.