I spent a few days on this piece, but lost interest after a while. It was tricky trying to get her to loosely wear the heels, as the model doesn't seem to understand my prompt. The heart-shaped ring holding the dress together was manually added.
Workflow: (1) Generate intial image with desired composition, (2) Use the depth-map ControlNet to generate better images, (3) Use Krita (w/ SD plugin) to refine the composition, (4) Perform Img2Img to get a refined output, (5) Upscale 2x with CCSR + HAT-L, (6) Further refine the details with Img2Img (30 steps @ 0.20 denoise), (7) Upscale 2x again, (8) Refine the image with even lower denoise, etc.
Prompt
seductive royal vampire, hand_on_thigh, slender_hands, red_nail_polish, detached_collar, golden_heart_ring_bra, sitting on chair, in luxurious fantasy castle, large_brick_walls, pointy_ears, toned_body, small breasts, big ass, amber red eyes, slit_pupils, happy, sexy, shallow depth of field, cinematic_lighting, Game_of_Thrones_throne, single_shoe, heel, blender_cgi_render, 3d, masterpiece, high_resolution, highres, best_quality, high_details
Negative prompt
lowres, missing_fingers, text, error, more_than_5fingers, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name
Vampire-chan!
I spent a few days on this piece, but lost interest after a while. It was tricky trying to get her to loosely wear the heels, as the model doesn't seem to understand my prompt. The heart-shaped ring holding the dress together was manually added.
Workflow: (1) Generate intial image with desired composition, (2) Use the depth-map ControlNet to generate better images, (3) Use Krita (w/ SD plugin) to refine the composition, (4) Perform Img2Img to get a refined output, (5) Upscale 2x with CCSR + HAT-L, (6) Further refine the details with Img2Img (30 steps @ 0.20 denoise), (7) Upscale 2x again, (8) Refine the image with even lower denoise, etc.