Glass Reality Weaver in Cosmic Hall

原文

GPT Image 2 · en

A grand, cinematic-quality digital art scene depicting a cosmic fantasy: In the center of the left side of the image sits a colossal humanoid figure composed of transparent glass and liquid crystal, its body filled with luminous internal filaments, refracted highlights, and intricate swirling textures beneath its surface. The figure's posture resembles that of a divine craftsman or a real weaver, its hands raised, stretching and kneading a massive, luminous golden veil on the right side of the image. The face is smooth and blurred, almost like a mannequin, devoid of any discernible expression, thus highlighting the sculptural form and the effect of light penetrating the translucent skin. This fabric of light, like a living tapestry or membrane, presents a complete fantasy world: a viaduct, a dazzling futuristic city with slender towers and domes, multi-tiered terraces, luminous buildings, and three clusters of deep green trees growing in the air or on high ground. The environment surrounding the colossal life form is a grand celestial hall, with ornate arches and feather-like branching filaments of golden light, opening onto the deep blue midnight sky. The ground, polished black mirror stone, clearly reflects figures, gold lines, and architecture. Five reflective metal or glass spheres of varying sizes are placed in the foreground, capturing the surrounding gold and blue light. The emphasis is on the dramatic refraction of the glass structures, the shimmering caustics, the intricate light trails, and the rich reflections on the glossy ground. Color palette: deep navy blue, black, silver-blue, and shimmering gold. Atmosphere: awe-inspiring, mysterious, sacred, and visionary. Highly complex, Baroque sci-fi fantasy aesthetics, wide-angle composition, epic scale, extremely sharp details, volumetric lighting, and surreal realism.

使い方

1. パラメータを含む元のプロンプト全体をコピーします。
2. 被写体、文字、構図を自分の要件に置き換えます。
3. 対応する外部ツールで使用し、利用条件と結果を確認します。

開始前の確認

• 参考画像の要否を確認してください。互換モデルの表示は、そのモデルでの生成実績を意味しません。
• 名前、数値、画像内の文字を確認します。モデルの版で結果は変わります。