Transform the provided 3D render into a modern comic book cover, using strict image-to-image style transfer, based exclusively on the visual language, linework, coloring, and graphic treatment of the provided reference comic image. They are standing, looking worried, on top of a building, with a dramatic night sky, lightning, and a full moon peeking through the clouds. The input 3D render will always change, but the comic book style must remain absolutely consistent with the reference image. Preserve with absolute fidelity everything from the original 3D render: – Character identities and silhouettes – Anatomy, proportions, facial features, and expressions – Hairstyles, skin tones, body types – Costumes, clothing layers, seams, folds, textures – Accessories, props, objects, weapons, tools – Damage, dirt, wear, imperfections if present – Poses, gestures, posture – Camera angle, framing, perspective, composition – Scene elements, background objects, and spatial relationships Do not invent, add, remove, modernize, reinterpret, stylize, simplify, exaggerate, or correct any element. Apply only a visual ressignification that converts the image into a comic book aesthetic: – Loose, imprecise, irregular lineart – Slightly dirty, sketchy, broken ink lines – Stylized cartoon-adult drawing, not realistic – Flat colors with limited palette – Hard, graphic shadows – Minimal to no gradients – Printed comic book texture feel – No 3D rendering traits visible The final image must look like a hand-drawn modern comic book panel, while remaining a one-to-one visual translation of the original 3D render. Nothing new may be introduced. Nothing may be removed. Nothing may be altered beyond style. photorealism, realism, hyperrealism, cinematic lighting, 3D render look, CGI, ray tracing, PBR materials, glossy surfaces, reflections, subsurface scattering, depth of field, motion blur, bloom, glow, lens flare, over-detailed textures, smooth clean lineart, manga style, anime style, watercolor, oil painting, painterly brushstrokes, airbrus no speech bubbles, no text, no logos, no onomatopoeia