Bilderthread AI Bilder

[KingPing]

Obiges Bild habe ich mal als Vorlage genommen, um mir daraus einen Prompt erstellen zu lassen. Dafür habe ich diese Webseite genutzt:

AI Image to Prompt Generator | Convert Image to Text Prompt Online Free | Zemith.com

Generate detailed text prompts from any image instantly with Zemith's free AI Image to Prompt Generator. Upload your photo, artwork, or design and get accurate, ready-to-use prompts for AI art, Midjourney, Stable Diffusion, and more. Fast, easy, and perfect for artists, students, and creators...

www.zemith.com

Der Prompt wurde recht umfangreich und detailliert und mein erstes Urteil zu dieser Seite fällt entsprechend, auch mit Blick auf das Resultat, ist rückblickend das Ergebnis toll. Den negativen Prompt lasse ich mal außen vor, weil da nur "no text, no watermark" drinsteht

A realistic studio photograph of a nude woman seated on a dark blue upholstered chair against a warm beige backdrop. The subject is a woman with long, dark brown hair that falls loosely over her shoulders, with a few wet strands clinging to her upper chest and right arm. She has a medium-tan complexion with smooth, luminous skin, dark almond-shaped eyes, defined dark eyebrows, and full lips with a subtle red tint. Her expression is neutral and direct, gazing straight into the camera with a calm, confident intensity. She is positioned centrally, sitting with her body angled slightly to her right, her left knee drawn up toward her chest while her right leg extends forward and down, her bare foot resting on the floor. Her left hand is raised near her chin, fingers lightly curled, while her right hand rests on her inner thigh near her lap. The lighting is dramatic and highly directional, coming from a large softbox positioned to the upper left, visible as a white glowing panel in the top left corner. This creates strong, warm highlights on the left side of her face, shoulder, and breasts, while casting deep, sculpted shadows on the right side of her body and face. The light source produces a distinct vertical striped pattern of light and shadow across her torso and thighs, likely from a grid or slatted modifier on the light. The background is a textured, draped fabric in a warm, muted ochre or tan tone, with soft folds and shadows that fall into darkness on the right side. The chair is a simple, modern design with a dark blue textured seat and backrest, supported by thin black metal legs. The image has a shallow depth of field, with the subject in sharp focus while the background remains slightly soft. The color grading is warm and cinematic, with rich golden-brown tones, deep shadows, and a slight vignette. The overall mood is intimate, artistic, and raw, capturing a fine-art nude study with an emphasis on form, light, and shadow. The composition is a medium shot from a slightly elevated angle, framed to include the full figure from the head down to the feet, with the lighting and pose creating a strong diagonal line across the frame. The image is rendered with high detail, natural skin texture, and professional studio lighting.

Hier der Workflow:

Anhang anzeigen 1790952442826.png

Das verwendete Model war dreamshaper_8.safetensors mit 1,99 GB Größe.

Anhang anzeigen 1790952481996.png

Das alles auf meinem ollen Laptop. Für mich ist ComfyUI halt Referenzklasse und es wird beständig verbessert. Erst diese Woche hat man sich von Version 0.37 verabschiedet und ist heute schon bei Version 0.38.1. Vielleich ein kleiner Sidestep dazu: Solltet ihr ComfyUI Desktop nutzen und die KI um Hilfe bitten, fügt bitte hinzu, dass deren Knowledge Base out of date ist und sie sich an die Online Dokumentation halten sollen.

Gleicher Prompt mit anderem Modell, realisticVisionV60B1_v51HyperVAE_pruned_fp16

Anhang anzeigen 1790954247132.png

 

[Maikel_elBlanco]

hier mal etwas Künstlerisches mit AI generiert...