GPT Image 2 prompts

@CreateImage
Generate a fresh and restrained ceremonial invitation visual around {argument name="theme" default="Good morning + Greeting + GOOD MORNING + City + Quote + Weather + 2026-09-01"}: The screen uses large areas of bright and clean white space as the main field. A crisp close-up image area is placed in the center, where the subject presents only the most emotional details and action relationships, allowing the viewer to read intimate, promising, or commemorative moments before the information. The main title uses {argument name="typography" default="relaxed and speedy handwritten strokes"}, with the font significantly enlarged, spanning across the upper edge of the image area and slightly overlapping, creating the first visual memory of text pressing onto the image. The title doesn't need to be neatly arranged; it should have the fluctuations, ligatures, thickness changes, and transparent overlays of real writing. A very small amount of fine information text can be placed on both sides, vertically close to the white boundaries for balance. Bottom information is arranged in stable fine print or small handwriting, with an extremely thin short line in the center for a pause. Colors are extracted from the theme's own materials and mood, keeping a {argument name="color scheme" default="Japanese and Korean style fresh color palette"} with large areas of white space to make the whole bright, light, clear, and quiet. Accent colors serve the emotional role of the title and key information. The overall finish should look like a designed private invitation, emphasizing the tension of the overlap between the image detail and the large handwritten title, sparse information rhythm, clean airiness, and a sense of ritual in the white space.

Subject:
Night Breeze Under the Tower

Main Subject:
The person is placed large in the center of the screen. A full-body night photograph of a {argument name="person" default="woman"} leaning against a {argument name="background element" default="white metal guardrail"} near a large radio tower lit in red-orange. Includes a group of traffic signs on the left, the tower in the back right, and the person brightened by a front flash all in one frame.

Character/Expression:
A small oval face with smooth cheeks and a rounded chin. Long brown eyes, natural straight eyebrows, a small straight nose, and coral lips with moderate thickness. Light pink blush and subtle eyeliner. A dark brown round bob ending below the chin, with thin bangs and short strands along the cheeks. Face turned toward the front, lips slightly open, looking quietly at the camera.

Clothing/Pose:
A short white fine-ribbed camisole with thin shoulder straps and a shallow round neck. Pale blue wrap-style shorts that overlap like a skirt, showing inner shorts. White ribbed socks and white sneakers. Leaning the waist against the center of the white guardrail, placing both hands on the left and right rails, legs wide open with one knee bent.

Background/Lighting:
Main light is a hard flash from the front of the camera. Black night sky, a large tower lit in white and red-orange in the back right, several red and blue traffic signs and a sticker-covered box on the left, distant cars, and streetlights. The hard frontal flash brightens the person and the rail, while the warm light from the tower colors the background.

Composition/Camera:
Low wide-angle perspective. Focus on the person, with the night view in a light blur. 3:4 vertical composition, a wide-angle full-body shot looking up from a very low position. Person in the center, traffic signs on the left edge, the tower in the top right, and open legs largely cut at the bottom left and right edges. The guardrail crosses behind the waist; keep the person sharp and the night view moderately blurred.

Texture/Style:
Photorealistic real-life photo. High-definition nighttime street photography utilizing direct flash. Captures white ribbed fabric, pale blue fabric, damaged metal rails, and the orange light of the tower in high contrast.

Negative:
Do not change the low wide-angle and the sitting posture on the guardrail; do not significantly change facial features.

{argument name="location" default="Aquarium"}
Taking a {argument name="composition" default="selfie in a tunnel-type tank"}.

Subject:
{argument name="subject" default="Blue Sarong on Stone Steps"}

Main Subject:
The person is placed slightly left of the center. A vertical photo at a {argument name="location" default="stone staircase beside an old ochre building"} of a blonde woman sitting sideways with her legs crossed. Capturing a {argument name="outfit" default="white cropped top and light blue floral sarong"}, and a white beaded bag on her left side under intense sunlight.

Person/Expression:
Slender oval face with a small chin, horizontally long dark brown eyes with upturned corners, and thin arched eyebrows. A thin bridge of the nose with a small tip, and plump glossy coral lips. Eyeliner extending from the outer corners and warm, light blush. Honey blonde loose wavy hair down to below the chest, parted near the center, with thin bangs falling onto the forehead. Tilting her head slightly to look at the camera, with lips slightly parted.

Clothing/Pose:
Wears a white camisole top and a blue wrap skirt. The off-white fine-ribbed camisole has thin shoulder straps, a deep V-neck, a ribbon at the center under the bust, and small ruffles at the hem. A thin light blue sarong with a white botanical pattern is tied at the waist, opened wide on one side to show the thigh. Sitting diagonally on the stone steps, overlapping both legs long toward the right foreground of the screen, with one hand on the knee and the other down on the step. A white woven beaded bag with a gold chain is placed on her left side.

Background/Lighting:
Stone steps on the left and an ochre wall on the right of the background, with the main light being hard direct sunlight from the upper left. Scuffed and mottled ochre plaster walls, coarse stone steps, a dark green door, brown flower pots, pale blue ceramic pots, and hanging vines. Hard sunlight from the upper left creates strong highlights on the hair, shoulders, and legs, casting sharp shadows on the wall and steps.

Composition/Camera:
The camera is at a knee-level portrait at the same height as the seat. The legs are large, extending toward the edge of the frame, with focus on the person. A 2:3 vertical composition, capturing from the same height as the seated person from the pots above to the feet. The person is at the left center, the extended legs are the large foreground at the bottom right, and the bag is at the bottom left. The legs continue off-screen, and the stone steps in the background stretch diagonally into the distance.

Texture/Style:
High-definition and warm photorealistic travel fashion photography. Depicting ribbed fabric, transparent sarong, woven beads, and old stone and plaster under strong natural light.

Negative:
Do not change the large opening of the blue sarong or the foreground scale of the long legs.

保護されたワードローブのルール — 最優先事項：以下の各衣服の説明を、不変の最終製品アイデンティティとして扱ってください。構造、パーツ、金具、素材、パターン、および登録されたすべてのグラフィックやマーキング（その表面、側面、向き、スケール、色、内容、スペル、数）を保持してください。最終的な着用状態は、対象となるアイテムの使用、位置、側面、向き、留め具、レイヤリング、折り目、タック、結び目、ドレープのみを上書きします。「ソース衣服」ラベルはスタイリング前のアイテムを表し、完成したシルエットではありません。保護された詳細を再設計、追加、削除、反転、複製、再配置、または描き直さないでください。ワードローブ：トップス：軽量で柔らかくシワ加工されたアイボリーのコットン製のクロップドオフショルダーブラウス。体にフィットする胴体と、密にギャザーを寄せたバストパネルが特徴です。ギャザーの入ったネックラインの上に深いV字型のフローラルレースインセットがあり、2つのレースの頂点から伸びる細いアイボリーのホルターコードが調節可能なタイを形成しています。伸縮性のある短い袖は柔らかいフリルを作り、フロント中央のドローストリングチャネルがバストにギャザーを寄せ、2本の細長いタイが垂れ下がっています。下端は幅広のアイボリーのアイレット刺繍バンド、フローラルカットワーク、スカラップヘムで仕上げられています。背面は伸縮性のある上端とアイレットバンドが続き、完全に対称的なクロップドシルエットになっています。レースのインセット、袖、タイ、アイレットヘムを一着の衣服として統合し、別のアンダーレイヤー、ロゴ、ラベル、テキストは含めないでください。ボトムス：中濃色のインディゴストレッチデニムを使用した、ローライズでフィット感のある5ポケットジーンズ。ヒップと太ももはタイトで、そのままクリーンなスリムレッグに続きます。等間隔のベルトループ、中央のジッパーフライ、円形のシルバーブルーのフロントボタンを備えたシェイプされたウエストバンドを作成してください。カーブしたフロントポケット、小さなコインポケット、外側のポケット角に合わせたシルバーブルーのアクセントリベット、ダークメタルの補強リベット、リアヨーク、クリーンなパッチポケットを追加してください。同系色のインディゴのステッチ、フロントライズの微妙なヒゲ加工、脚上部のソフトな縦落ち、そして綺麗に仕上げられた裾を使用してください。ベルト、プリントグラフィック、ロゴ、読み取り可能なテキストは含めないでください。フットウェア：解剖学的にミラーリングされた一対の、ダスティスカイブルーのキャンバス製ローカットレースアップスニーカー。局所的なミントアクアのグラフィックとアイボリーのラバーが特徴です。各シューズはコンパクトなロープロファイル、幅広の丸みを帯びたクローズドトゥ、そしてわずかにかかとが上がった厚みのあるほぼ水平なプラットフォームを備えています。パッド入りの低い履き口が長いシュータンを囲んでいます。アッパーは、ブルーのキャンバスヴァンプ、対向するアイレットフェイシング、幅広のクオーターパネル、補強されたヒールパネルを組み合わせています。アイボリーのラバーシェルトゥキャップと深いアイボリーのフォキシングサイドウォールが下端を縁取っています。外側のクオーターには、ギザギザの葉、2本のカールした茎、細かい脈のハッチング、ドットのアクセントがある、密度の高いミントアクアのアネモネ3輪のスプレーが描かれています。「IVY ARC」という大文字の単語が、アイボリーのフォキシング全周に沿って、1単語3文字の等間隔の単一行ユニットとして広いトラッキングで連続して繰り返されています。銀色の小さなアイレットが偶数列に並び、平らなアイボリーの靴紐が通常の十字路を通っており、紐は中央でコンパクトなリボン結びになっています。内側のクオーターは無地のブルーキャンバスのままで、リアパネル、アイボリーの裏地、ダークなトレッドは、解剖学的な反転を除いて左右で一致しています。キャンバスはマットに、ラバーはソフトな質感に、植物のラインワークは鮮明に、そしてすべての浮き出し文字を明確に判読できるようにしてください。バッグ：滑らかな連続フォイル表面、しっかりとした形状保持、丸みを帯びたエッジパイピング、ポリッシュ仕上げのシルバー金具を備えた、高光沢メタリックコーティングのレザー風素材で作られたコンパクトなメタリックシルバーのミニトート。ボディは、上部の角がわずかに丸みを帯びた短いほぼ長方形の封筒型で、スリムなサイドガセットと平らな補強ベースがあります。ハンドル間に狭いトップアクセス開口部があります。フロント、リア、ガセット、ベースを、境界エッジの周りに連続したパイピングを持つ一つの構造化されたボディとして構築してください。2つの短いアーチ型ハンドルを、細いステッチタブでフロントとリアのアッパーフェイスに取り付けてください。アッパーサイドシームの小さなリングに固定されたクリップ末端を持つ、細長いお揃いのストラップを1本追加し、長方形の長さ調節スライダーをはっきりと見せてください。上部の開口部は、コンパクトなジッパーと目立たないポリッシュ仕上げのスライダーで閉じてください。フロント、リア、ガセット、ベースには追加のポケットやグラフィックを入れず、ストラップアンカー、ハンドルタブ、パイピング、ジッパー、ベース補強が同じボディに物理的に結合されているようにしてください。鮮明な構造、制御された継ぎ目の厚み、滑らかなメタリックのカバレッジ、そしてソフトな指向性反射を維持してください。最終着用状態：視点条件付きボディサイドマップ：正面図 — 着用者の左は画像の右、着用者の右は画像の左。背面図 — 着用者の左は画像の左、着用者の右は画像の右。側面図、交差した手足、または曖昧な場合は、対応する肩、腰、目、または耳の目印から名前の付いた解剖学的側面を連続的に追跡してください。この状態は指定された解剖学的側面にのみ適用し、反転、入れ替え、または複製は決してしないでください。- バッグの持ち方：最終状態：登録されたねじれのないストラップまたはチェーン1本が、元の両端で取り付けられ、着用者の右肩からすべての重量を支え、解剖学的なフロントセンターラインを一度横切り、着用者の左前の腰でバッグを平らに支えます。バッグの背面は体に向き、前面は外側を向きます。両手は離したままにします。登録されたサポートとコンポーネントを接続したままにし、選択したルートに沿ってのみ接触と重力による余裕を許容してください。この操作内での自然な取り扱い：選択したルートとすべての取り付けポイントを正確に保ちながら、素材に応じた適度な緩みと自然な体との接触を許容してください。

Subject:
{argument name="subject" default="Backstreet Cargo Style"}

Main Subject:
The person is placed large in the center of the screen. A full-body fashion photograph of a woman walking with both hands in her pockets through a {argument name="background" default="narrow residential street with sunlight filtering in"}. The main focus is on the {argument name="outfit" default="white cropped camisole and extremely loose light blue denim cargo pants"}.

Person/Expression:
Slightly slender oval face with a small, not too pointed chin, horizontally long grayish-brown eyes with slightly upturned corners, and thin arched eyebrows. A thin bridge of the nose with a small tip, thin upper lip and slightly thick coral lower lip. Light pink blush and thin eyeliner. Light brown loose wavy hair down to below the chest with a deep side part, leaving thin bangs and strands around the face. A quiet expression looking straight at the camera.

Clothing/Pose:
A short white camisole with a fine pattern has ultra-thin shoulder straps and a deep round neck. The light blue, very wide denim cargo pants feature drawcords, large and small 3D pockets, raw edges, and numerous hanging straps. White platform sneakers, large gold hoop earrings, and a small beige chain bag on the right shoulder. Walking with both hands in the pants pockets and one foot stepped forward.

Background/Lighting:
Plants on the left and a house on the right of the background, with the main light being hard direct sunlight from the upper left. Low-rise houses, utility poles and wires, dark green plants on the left, a red mailbox in the back right, and a narrow asphalt road. Strong sunlight from the upper left edges the hair and white top, casting a long shadow of the person onto the road surface.

Composition/Camera:
The camera is at a front eye-level at waist height. The shoes are contained within the frame, with focus on the person and light bokeh in the background. A 3:4 vertical composition, a full-body photo from waist height. The person is placed large in the center, with the vanishing point of the alley behind the head, emphasizing the wide pants and white sneakers in the lower half. The background is moderately blurred to make the pockets and straps of the outfit vivid.

Texture/Style:
High-definition photorealistic street fashion photography. Depicting white fabric, faded denim, fraying, metal chains, and asphalt with cool blue-grays and natural skin tones.

Negative:
Do not turn ultra-wide multi-pocket cargos into skinny pants; no major changes to facial features.

A photorealistic night time car selfie of a {argument name="subject" default="young woman"} sitting in the driver’s seat, turning slightly toward the camera with a playful wink and a soft smile. She has long straight blonde hair with delicate thin braids framing both sides of her face, decorated with tiny white beads. She wears an {argument name="outfit" default="oversized black hoodie and light-wash ripped jeans"}, with a subtle silver nose ring. The interior of the {argument name="car type" default="luxury car"} features black leather seats, a modern digital dashboard, illuminated instrument cluster, central touchscreen, and elegant purple ambient LED lighting running across the dashboard and center console. Outside the windows, a city street at night is visible with colorful blurred lights and strong bokeh. Candid smartphone photography, realistic facial features, natural skin texture, detailed hair and braids, soft low-light exposure, cinematic nighttime atmosphere, subtle film grain, shallow depth of field, realistic reflections on glass, cozy luxury-car aesthetic, high detail, vertical composition, 4:5 aspect ratio.

Create an ultra-realistic 8K cinematic photograph of the {argument name="reference person" default="girl from the uploaded reference image"}, preserving her exact facial identity, natural skin tone, facial structure, hairstyle, body proportions, and all recognizable features with maximum accuracy. Capture her from an {argument name="perspective" default="extremely low, ground-level perspective"} using a dramatic {argument name="lens" default="ultra-wide fisheye lens"}, creating a bold and immersive sense of depth while keeping her face and overall appearance natural and realistic. The composition should feel like an authentic high-end photograph captured with a professional camera, with realistic perspective distortion around the edges and subtle fisheye curvature. Use natural, cinematic lighting with soft highlights and realistic shadows that complement her face and skin texture. Preserve fine details such as individual strands of hair, natural pores, subtle facial texture, realistic eyes, lips, and delicate imperfections. Avoid excessive beauty retouching, artificial smoothness, plastic-looking skin, or exaggerated facia…

Ultra-realistic candid full-body street-style photograph of the same young woman from the reference image, preserving her recognizable identity, facial features, proportions, olive-green eyes, deep brunette hair, natural freckles, skin tone, and overall appearance with high fidelity.

She is casually standing inside a {argument name="location" default="modern neighborhood grocery store"} near the entrance, caught in an unposed editorial moment as she glances slightly away from the camera with a relaxed, confident expression. She wears a {argument name="dress color" default="burgundy-red"} sleeveless mini dress with a subtle square neckline, sheer black ankle socks, and stylish black-and-white animal-print heeled sandals. A compact patterned shoulder bag hangs naturally from one shoulder, while a lightweight black plaid jacket is loosely draped over her forearm.

Her deep brunette hair is gathered into an effortless messy low updo, with a few loose strands naturally framing her face. Keep her natural freckles, realistic skin texture, subtle pores, and visible tattoos on her arms.

The environment feels authentic and lived-in: polished tiled floor, stacked grocery shelves, glass-door beverage coolers, colorful product packaging, handwritten promotional signs, fluorescent ceiling lights, and a partially visible storefront entrance. Include subtle reflections from the refrigerator glass and realistic background activity without distracting from her.

Candid fashion photography, eye-level perspective, natural body proportions, slightly imperfect framing, realistic ambient lighting, soft film grain, muted highlights, authentic smartphone-camera feel, shallow but believable depth of field, no beauty filter, no CGI appearance, photorealistic 4K detail.

Create a premium flagship commercial poster for an original compact camera brand named "LUXAR", featuring a hero model called "VISTA M". Use ultra-realistic macro wildlife-product photography with a print-ready luxury electronics advertising finish. The core brand super-symbol must be a living animal profile whose eye is replaced by a perfectly integrated camera lens, turning adaptive vision into a physical product metaphor. Core composition: preserve the exact structural logic of a vertical poster with three zones. Upper 70%: an extreme macro side-profile of a chameleon facing left, tightly cropped so the snout enters from left-center and the body extends toward the right edge. The eye area is replaced by a realistic front-facing camera lens, embedded seamlessly into the reptile anatomy. Upper-left: a bold brand title block with a red logotype, one short white tagline, and four white corner brackets framing the title zone. Lower 25%: three front-facing compact cameras aligned horizontally on a pale reflective surface. Bottom 18%: a black technical information band with the model name large at center-right and smaller feature text at left and right. Port logic: turn this exact visual system into the unmistakable brand super-symbol of LUXAR. The permanent identity must be built from four fixed codes: a macro animal profile, a camera lens replacing the eye, a three-color product lineup at the bottom, and a black technical spec band grounding the ad. The image should feel instantly ownable, repeatable, and scalable across a full campaign series where different animals embody different imaging technologies. Orbit logic: build a strong closed visual loop across the poster. The eye should move from the upper-left title block into the lens-eye, follow the chameleon snout and jawline, sweep down into the three product cameras, lock onto their aligned lens centers, travel across the black spec band, and rise back toward the embedded lens through the animal’s neck contour and background blur. The lens highlight, facial profile, camera lineup spacing, and typography blocks must all reinforce one continuous circulation path. Hero metaphor: the chameleon must embody adaptive optics, responsive focus, and visual precision. The lens-eye integration must feel mechanically plausible and biologically fused without gore, distortion, or horror. The metaphor should read instantly, cleanly, and commercially. Animal detail: render the chameleon with dense micro-scale texture, green-turquoise-iridescent scale breakup, darker ridge segmentation, realistic skin dryness, tiny bumps, natural jaw serration, and believable cranial volume. The eye socket anatomy must transition cleanly into the lens housing. The creature should feel still, alert, and physically real. Lens detail: the embedded lens must be highly realistic, with concentric glass elements, black metal barrel, subtle engraved technical markings, purple-green coating reflections, visible aperture depth, and crisp circular geometry. The lens should be proportioned naturally to the chameleon head while remaining immediately legible as premium optics. Lighting: soft directional macro lighting from upper-left front, medium-soft quality, with controlled highlight separation across scale texture, lens rim, and facial contours. Secondary fill from front-right at lower intensity to preserve shadow-side detail. Product cameras below should receive cleaner studio-style frontal lighting with crisp highlight bands and restrained tabletop reflections. Contrast should be moderate-strong, with readable midtones and no crushed blacks. Color system: dominant muted olive-beige background, green-to-turquoise chameleon skin, black lens core, red brand accent in the upper-left, and three product body colors below: deep cobalt blue, satin black, and dark metallic crimson. Bottom spec band black with white and red text accents. The palette must feel natural-tech, restrained, and product-led. Product showcase: display three front-facing compact cameras on the lower strip, evenly spaced, same model, identical lens size, with body colors blue, black, and burgundy-red. Cameras must have machined edges, matte-metal shells, realistic lens glass, minimal branding, and premium compact proportions. Add subtle mirrored reflections directly below each camera. Typography and layout: top-left title block: LUXAR See Beyond Focus Use white corner brackets around this title zone. Bottom black band: large model name: VISTA M left feature stack: Touchscreen OLED Creative Filters 4K Motion Capture Interchangeable Lens Ready 24 MP Sensor ISO 100-12800 right paragraph: a short original line about adaptive imaging, compact engineering, and fast visual response. All wording must be original English only. Materials and texture: extreme sharpness on reptile scales, lens housing, glass reflections, product camera edges, and typography. Preserve realistic reptile skin roughness, subtitle dusting on scale edges, matte painted camera bodies, glossy lens glass, and clean print-band contrast. Everything must obey real optical and lighting physics. Depth of field: very shallow macro depth of field in the upper animal zone, with the lens-eye and nearest facial scales critically sharp while background branches dissolve smoothly. Lower product row must remain fully sharp in clean studio-product style. Image quality and physical rules: high-resolution commercial clarity, no muddy blur on the hero lens, no broken anatomy, no extra limbs, no warped reptile face, no floating cameras, no messy reflections, no black blotches, no unreadable text, no random props, no copied real brand marks, no watermark. Negative prompt: real Canon logo, copied slogan, copied model name, cartoon reptile, horror mutation, gore, broken jaw, deformed scales, warped lens, unreadable text, low-detail camera body, low-resolution macro, muddy shadows, dead black banding, messy layout, random extra products, plastic-looking skin, illustration style, watermark.

Authentic news documentary photography, 2017 Tobu Zoo, Saitama Prefecture, Japan; {argument name="subject" default="Real Humboldt penguin Grape-kun"}, adult male, black and white feathers, purple identification ring naturally visible; realistic penguin enclosure, pool, rocks, metal fencing, trees, visitors; "Kemono Friends" character Hululu cardboard cutout, realistic cardboard material and base; natural ambient light, candid snapshot, realistic feather texture, slight natural grain, low saturation, natural life feeling, restrained, gentle.

[Image 1 | April 2017]
During the collaboration between "Kemono Friends" and Tobu Zoo, the Hululu cutout appears in the penguin area; Grape-kun naturally appears near the cutout; visitors watch and take photos through the fence; authentic zoo atmosphere.
Text:
"April 2017"
"Tobu Zoo x Kemono Friends"
"Hululu's cutout appears in the penguin enclosure."

[Image 2 | Gazing]
Grape-kun stands quietly near the Hululu cutout, body naturally facing the cutout and gazing continuously; depicts the behavior noted by visitors and zoo staff at the time; no anthropomorphism, no expression of human-style love; background naturally blurred like an accidentally captured news photo.
Text:
"Grape-kun seems to particularly care about this cutout."
"'Love' is the interpretation people give to its behavior."

[Image 3 | Staying]
After the collaboration ends, other character cutouts are removed, but the Hululu cutout remains in the penguin area; other positions left naturally vacant; Grape-kun still appears near the cutout; quiet penguin enclosure, natural soft light.
Text:
"The collaboration has ended."
"Other cutouts were removed, but Hululu stayed."

[Image 4 | Farewell]
The time point after Grape-kun's death on October 12, 2017; the penguin enclosure is quiet and empty, Grape-kun is gone, the Hululu cutout still exists; pool, rocks, fence, and empty activity space; soft natural light, expressing absence through space, no death scene shown.
Text:
"October 12, 2017"
"Grape-kun passed away at age 21."
"We cannot know what it truly saw that cutout as."

Strictly based on real events, no fictionalization of Grape-kun's psychology, dialogue, memories, or motives; "falling in love with 2D" and "liking Hululu" are human interpretations of its behavior.

Keep "● DeepBlue" in the bottom left corner of each image; the dot is #0B3D91 deep blue, and "DeepBlue" is a white natural handwritten font.

Please turn each photo I upload into an individual high-end design poster, rather than a collage of multiple photos, outputting each photo separately. The overall composition should be a 3:4 vertical layout, with the top and bottom areas strictly maintaining a 1:1 height ratio, each occupying 50% of the frame.

The upper half should retain the original photo, preserving the identity of the subject, structure, posture, authentic texture, natural lighting, and original color atmosphere, with only slight high-end color grading to give it the texture of an art magazine, independent publication, or exhibition image. The background can be naturally extended to fit the aspect ratio, but the subject must not be stretched, distorted, or changed.

The lower half should not be a complete replica of the photo, nor a simple downsized redrawing of it. Instead, it should extract the most memorable subjects, actions, partial outlines, objects, plants, architectural fragments, color relationships, and symbolic details from the original image, re-refining them into several visual elements converted into a unified style. These elements, combined with text, will form a Rebus-style visual sentence. These images are not decorative illustrations; they must truly replace words in the sentence. Once a word is replaced by an image, the word must not be written in text again. The entire sentence must rely on the combination of 'text + images' to be fully understood.

First, generate a short, imaginative, and slightly poetic or editorial sentence based on the photo's mood, action, relationships, location, memory, or metaphor. Then, select several nouns, action words, imagery words, or emotion words from the sentence that are best suited for visualization and replace them directly with corresponding image elements. For example, flowers, windows, birds, trees, cups, the sea, wind, lamps, roads, waiting, passing, returning, and early morning can all be represented by visual elements derived from the original photo. The images must occupy the original position of the words, forming a truly continuous 'text → image-word → text → image-word → text' reading logic, rather than just small drawings decorating a complete sentence.

Each replaced 'visual word' should be placed within an independent, small colored block background, like a small fragment cut from the memory of the original photo. Each block can have different sizes, aspect ratios, and cropping styles—nearly square, horizontal strips, vertical blocks, or small irregular cutouts. The blocks do not need to be uniform in size but must have a clear rhythm and hierarchy. The images within these blocks are not screenshots of the original photo but artistic fragments redrawn in a unified visual language, retaining only the most recognizable outlines, actions, structures, and essence. They can be partially cropped, showing only a portion, or naturally extend beyond the block's edges, making them look more like a meaningful visual word rather than a standard illustration.

These visual elements should overall adopt a unified modern hand-drawn editorial illustration style: reconstructing original elements through simplified outlines, generalized forms, a few flat color blocks, and a slight handmade feel, avoiding excessive realism, complex volumes, photographic textures, and mechanical reproduction. The images should retain the soul and recognizability of the original subject but not strive for complete depiction, feeling more like refined visual memories. The style should be light, clever, design-oriented, and reminiscent of art publications, ensuring all image-words are unified and harmonious.

The color block backgrounds behind each visual word should follow a 'fresh, bright, and eye-pleasing' soft color logic, though each block can have a different color. Colors should be extracted from the brightest, most vibrant, and most vital colors in the upper photo and then retuned, rather than simply taking an average. Increase the overall brightness and color purity to create a softer, lighter, and more approachable color palette, naturally expressing hues like milk blue, lake cyan, mint green, sprout green, cream yellow, peach pink, cherry blossom pink, apricot orange, and lavender purple. There can be slight shifts between blocks, but the overall look must be unified, comfortable, transparent, and clean, conveying a relaxed, pleasant, healing, slightly feminine, and 'Instagram-style' aesthetic. Avoid dirty grays, dark tones, aged effects, heavy Morandi palettes, fluorescent colors, cheap candy-like vibes, and excessive contrast.

The composition should be organized around the 'reading path of the sentence,' making the bottom half look like a page of a visual poem. Text and image-words can be spread across 2–4 lines; absolute alignment is not required, but there must be a clear, elegant, and natural reading rhythm. The size of the image-words should vary according to their semantic importance: some may be the size of a regular word, while others can be slightly larger to act as visual focal points. Use varying heights, density, slight offsets, white space, and baseline rhythm between text, image-words, and color blocks to create an asymmetrical but stable editorial layout. Avoid sticker-wall, card-list, regular grid, or messy collage appearances.

The bottom half should retain a significant amount of white space, allowing the void itself to become an essential part of the breath, rhythm, and sophisticated feel. The background should be near-white, off-white, or an extremely light shade that intelligently coordinates with the overall color temperature of the original photo, remaining clean, quiet, and transparent. All image-words, text, and color blocks should appear suspended within ample white space, making the layout resemble an art magazine, independent publication, visual poem page, or experimental graphic work rather than a commercial poster template.

There is no restriction on the language of the text; you may freely choose Chinese, English, bilingual, or any language that suits the atmosphere of the photo, provided the content is concise, natural, and imaginative. Avoid common commercial headline fonts; instead, use light, fresh, delicate, and thin fonts with a modern editorial feel and a slight humanistic touch. Character spacing should be naturally relaxed with plenty of white space and a restrained yet dynamic layout. Slight offsets, line breaks, and rhythmic changes are allowed so that the text and image-words coexist as equals, jointly forming the image rather than one dominating the other.

The overall visual effect should be: the top half is the realistic image, and the bottom half is a visual sentence composed of 'text + image-words + color blocks.' It should look like an emotional short poem, a memory fragment, or a visual metaphor extracted from the original photo, reorganized into a light, clever, fresh, healing, modern, and design-forward piece of art. The focus is not on describing what is in the photo, but on translating the photo's most moving elements into a graphic language that can be both read and viewed.

Avoid: directly using original photo crops as visual words, treating images merely as decorative illustrations next to text, including images without deleting the corresponding words in the sentence, sticker-wall aesthetics, card-grid layouts, commercial template styles, information-stream layouts, e-commerce vibes, childlike collage feels, messy disorder, muddy colors, cheap candy colors, excessive realism, complex backgrounds, mechanical fonts, and crowded layouts lacking white space.

Create a cozy, cinematic anime-inspired lifestyle portrait of a cute young woman with a {argument name="hair style" default="short pastel-pink bob haircut"}, wearing an {argument name="shirt" default="oversized vintage red plaid flannel shirt"}, relaxed baggy blue jeans, and classic canvas sneakers. Beside her is a {argument name="cat breed" default="fluffy gray-and-white British Shorthair cat"} wearing a tiny matching plaid bandana.

Capture a warm, candid moment between them during golden hour. They should interact naturally and affectionately, with soft sunlight falling across the scene and creating long, gentle shadows on a clean minimalist beige wall and ground.

Add a whimsical magical twist: transform their realistic shadows into playful hand-drawn manga illustrations. The woman's shadow becomes an adorable chibi anime version of herself in a joyful, energetic pose, while the cat's shadow transforms into a funny cartoon version of the cat, playfully waving, dancing, or reaching toward her.

The illustrated characters should appear to emerge naturally from the real shadows, blending seamlessly with the environment. Incorporate delicate pencil-sketch linework, expressive manga motion marks, tiny hearts, stars, butterflies, and spontaneous doodles organically within the shadow shapes.

Visual style: cozy slice-of-life anime blended with premium editorial photography, minimalist composition, warm cinematic golden-hour lighting, authentic candid emotion, dreamy yet believable atmosphere.

Quality & realism: highly detailed realistic fur, natural skin texture, authentic fabric and clothing details, subtle environmental textures, physically believable sunlight and shadows, soft atmospheric depth, shallow depth of field, 50mm photography look, Kodak Portra 400-inspired color rendering, photorealistic subjects contrasted with hand-drawn manga shadow art, sophisticated visual balance, ultra-detailed, cinematic, magical yet believable, masterpiece quality, 8K.

Realistic portrait photography, vertical; adult East Asian female, fair and delicate real skin; long hair, Song Dynasty style low bun, a few strands of hair falling naturally on the side of the face; {argument name="glasses" default="rimless thin metal frame glasses"}; exquisite classical makeup, slender eyeliner, long eyelashes, light pink blush, vermilion lip color. Wearing {argument name="outfit color" default="hazy blue"} Song Dynasty style Hanfu, the upper jacket is made of delicate dark-patterned fabric, with a light beige outer robe, wide sleeves falling naturally, narrow waistband, paired with simple jade earrings and a plain hairpin, overall elegant and restrained. The character sits {argument name="setting" default="by a wooden couch near the window"}, with hazy bamboo shadows outside the window, fingers lightly resting on the sleeves, body naturally tilting towards the camera, quiet posture. Looking directly at the camera with half-closed eyes, brows slightly furrowed, eyes cold and disgusted, with a touch of disdain and scrutiny; lips naturally slightly open, revealing a small amount of upper teeth. Hazy blue, off-white, and wood tones form a low-saturation Oriental color palette, with real fabric texture, hair strands, and skin details, soft and natural.

A stylish alternative rock aesthetic photoshoot of a young woman sitting on the floor indoors holding a {argument name="guitar" default="red electric guitar"}, dramatic candid pose with her head tilted back and tongue playfully out toward the guitar neck, edgy playful attitude, {argument name="hair style" default="black hair with subtle colored streaks"}, wearing {argument name="outfit" default="oversized luxury streetwear"} and patterned designer pants, lots of rings and jewelry on her fingers, studio flash photography, direct on-camera flash, high contrast lighting, sharp focus, realistic skin texture, shallow depth of field, modern editorial music photography, clean background, fashion + rock vibe, 35mm lens look, ultra high detail.

A photorealistic night time car selfie of a young woman sitting in the driver’s seat, turning slightly toward the camera with a playful wink and a soft smile. She has {argument name="hair style" default="long straight blonde hair with delicate thin braids"} framing both sides of her face, decorated with tiny white beads. She wears an {argument name="outfit" default="oversized black hoodie"} and light-wash ripped jeans, with a subtle silver nose ring.

The interior of the luxury car features black leather seats, a modern digital dashboard, illuminated instrument cluster, central touchscreen, and {argument name="lighting" default="elegant purple ambient LED lighting"} running across the dashboard and center console. Outside the windows, a city street at night is visible with colorful blurred lights and strong bokeh.

Candid smartphone photography, realistic facial features, natural skin texture, detailed hair and braids, soft low-light exposure, cinematic nighttime atmosphere, subtle film grain, shallow depth of field, realistic reflections on glass, cozy luxury-car aesthetic, high detail, vertical composition, 4:5 aspect ratio.

Create a single vertical 9:16 pre-production development board combining {argument name="elements" default="CHARACTER SHEETS + CINEMATIC STORYBOARD"} for [subject]

Create a premium, modern high-fashion CHARACTER CONCEPT ART SHEET in a 16:9 widescreen layout on a pure white background. THE ATTACHED REFERENCE IMAGE DEFINES THE ART STYLE — treat it as a strict style bible and reproduce its exact rendering

[iPhone 18 Preview | Apple Keynote Series]

[01 | First Foldable iPhone] English title: THE FIRST FOLDABLE iPHONE. Main visual uses iPhone Fold CAD and dummy unit cues: passport-style wide body, short and wide when folded; nearly 4:3 large inner screen when unfolded; ultra-thin body; central precision hinge; horizontal dual-camera platform; white or silver metal glass body; semi-unfolded state. Realistic high-end product photography. Large white space at the top, product is huge and located bottom center.

[02 | A20 Pro] English title: 2nm. A20 PRO. Huge A20 Pro chip as the sole visual subject; realistic advanced semiconductor packaging, metal contacts, microstructure, natural reflection. Pure white or light silver-gray background, large white space, realistic macro product photography.

[03 | Camera Upgrade] English title: THE CAMERA CHANGES. Huge iPhone 18 Pro main camera macro close-up as the sole visual subject; internal mechanical variable aperture blades clearly visible inside the lens. Precision metal structure, real glass lenses, natural optical reflections. White or light silver-gray background, high-end product photography.

Create a premium vertical editorial fashion portrait with a surreal double-exposure artistic composition. A young woman walks confidently toward the camera in a full-body pose, hands in the front pockets of high-waisted black jeans, wearing a fitted plain black short-sleeve T-shirt and white sneakers. She has {argument name="hair style" default="long, voluminous, slightly messy wavy dark brown hair with soft bangs"}, delicate youthful facial features, warm light-to-medium skin, subtle natural makeup, and {argument name="eyewear" default="thin round wire-frame glasses"}. Behind and above her, blend a much larger close-up portrait of the same woman into the background, showing her face and shoulders with a calm, gentle expression, looking directly at the viewer. The composition contains exactly two depictions of the same woman: one large semi-transparent close-up head-and-shoulders portrait in the upper background and one smaller full-body walking figure centered in the foreground. Surround both figures with dramatic watercolor ink clouds, splashes, drips, and smoky textures in {argument name="main ink color" default="deep emerald green"}, black, charcoal gray, and white, spreading outward like an abstract painted aura. Use an off-white textured paper background with visible grain, irregular ink blotches, vertical paint runs, and scattered splatter marks. Add a glossy reflective wet floor at the bottom, mirroring the green and black ink and the lower body subtly. Lighting should feel cinematic and editorial, with soft sunlight highlights on the hair, realistic skin texture, sharp facial detail, and a premium magazine-cover finish. Keep the subject centered, vertical 9:16 framing, full-body figure from head to shoes visible, large portrait cropped near the top edge, no text, no logo, no watermark.

Create a tall 1:3 vertical cinematic portrait of {argument name="subject" default="a beautiful young East Asian woman"} reclining sideways on an antique floral upholstered sofa, her head resting on folded hands against the sofa back as she gazes softly toward the camera. She wears {argument name="outfit" default="a sleeveless pale silk qipao-style dress with muted watercolor floral patterns, a high collar, and a thigh-high side slit"}; her dark hair is styled in a loose low updo with soft face-framing strands, with delicate earrings and natural romantic makeup. The pose is elegant and intimate: upper body curled near the sofa arm, one bare shoulder visible, the long dress flowing down the frame, one leg partially visible through the slit in the lower foreground. Use {argument name="lighting style" default="warm diffused window light with hazy bloom"}, heavy soft focus, shallow depth of field, dreamy film grain, muted beige and cream tones, and a nostalgic boudoir editorial mood. Compose with the face in the upper-left third, the body extending diagonally downward through the very narrow frame, an overexposed white glow along the right edge like light leaking through a curtain, and blurred dark foreground vignetting on the left. Background should be minimal: a dim beige wall, soft shadows, and vintage fabric texture. Photorealistic, 85mm portrait lens feel, f/1.4 depth of field, painterly softness, no text, no watermark, no extra people.

Goal: Create a premium cinematic action storyboard collage for a short Hollywood-style fight sequence starring {argument name="character name" default="a rugged dark-haired South Asian male action hero"}.

Canvas: Vertical 3:4 poster-like collage, approximately 900 x 1200 px, divided by thin white borders into exactly 7 cinematic panels. Use a dark, gritty, photorealistic IMAX blockbuster look with high contrast, shallow depth of field, smoky warehouse atmosphere, wet-black wardrobe highlights, dramatic overhead industrial lighting, and realistic motion blur.

Setting: A large abandoned industrial warehouse at night with concrete floors, steel beams, metal shelving, barrels, crates, ladders, haze, dust, sparks, and harsh rectangular lights in the background. The overall color grade is charcoal black, steel gray, muted brown, and cool white highlights.

Main character: The same male hero appears consistently in every panel: early-to-mid 30s, athletic build, medium tan skin, intense brown eyes, thick dark eyebrows, black slightly messy medium-length hair falling over the forehead, short full beard and mustache, sharp jawline, serious expression. He wears a fitted black tactical T-shirt, black leather or tactical jacket, black pants, and black belt throughout. Do not change his face, hair, age, or wardrobe between panels.

Panel layout and exact content count: Include exactly 7 panels total:
1. Top wide panel: The hero stands in the center of a tense standoff, facing three male attackers in dark jackets. One attacker is seen from behind on the far left foreground, one stands behind the hero on the left, and one faces him on the right foreground. The hero looks focused and ready to fight.
2. Second wide panel: Extreme close-up portrait of the hero's face, filling most of the frame, intense direct eye contact, damp hair strands over forehead, blurred dark background.
3. Third row left angled panel: The hero throws a powerful punch or elbow strike at an attacker, with sparks, dust, motion blur, and the attacker recoiling.
4. Third row right angled panel: The hero is surrounded by three attackers in a dynamic fighting stance, arms extended, ready to counter multiple incoming strikes.
5. Fourth wide panel: The hero kneels over a defeated attacker on the warehouse floor, one hand pinning him down, while other unconscious bodies lie in the background.
6. Bottom left panel: The hero walks forward confidently down the warehouse aisle, adjusting or gripping his black jacket with both hands, defeated men lying on the floor behind him.
7. Bottom right panel: Tight dramatic close-up of the hero's face, cropped on one side, intense eyes and serious expression, dark blurred background.

Action details: Show exactly 3 main attackers in the confrontation and group-fight panels, all adult men in dark clothing. Across the later panels, show multiple defeated bodies on the floor but keep them secondary and slightly blurred. The mood should suggest a fast, brutal 15-second compressed fight montage rather than a static portrait.

Visual style: Photorealistic live-action film stills, AAA action movie production value, ultra-realistic skin texture, cinematic lighting, high dynamic range, anamorphic depth of field, subtle film grain, no comic or illustration style. Use crisp facial details in close-ups and controlled motion blur in fight panels.

Text and constraints: No captions, no logos, no watermark, no UI elements. Keep the collage borders clean and white. Maintain the same hero identity and outfit in all 7 panels. The atmosphere should feel explosive, aggressive, and continuously moving.

Create an ultra-realistic full-body studio portrait showing exactly two versions of the same {argument name="character identity" default="adult Black man"} standing side by side against a seamless warm beige photography backdrop. On the left is the full-size realistic human version: tall, stylish, confident, short neatly trimmed black hair, smooth clean-shaven face, wearing black angular sunglasses, looking downward toward the smaller version with a calm cool expression. On the right is one miniature stylized 3D cartoon version of the same man, about one third of the human's height, with an oversized head, large expressive eyes, rounded youthful facial features, short black hair, and a charming confident pose, looking up at the human version. Both figures wear the same outfit: a deep {argument name="shirt color" default="navy blue"} long-sleeve button-down shirt with sleeves rolled to the forearms, {argument name="trouser color" default="camel brown"} tailored dress pants, a brown leather belt with metal buckle, brown polished dress shoes, and a gold wristwatch. The human stands with both hands in his pockets, feet slightly apart; the miniature cartoon also stands with hands in pockets, mirroring the pose. Use professional studio lighting with soft shadows on the floor, warm tan gradient background, high-fashion editorial composition, crisp realistic fabric textures on the human, glossy high-quality Pixar-like 3D rendering for the miniature, harmonious scale contrast, vertical 2:3 framing, full bodies visible from head to shoes, no text, no logos, no extra people, no props.

Create a photorealistic vertical beach portrait of a young {argument name="subject description" default="East Asian woman in her early twenties"} standing by a vivid turquoise ocean under a bright blue summer sky. She is shown from the waist up in a three-quarter back pose, turning her head toward the camera with a soft confident half-smile and relaxed eyes. She wears a wide-brim natural straw sunhat with open woven texture and a dark navy rim; one hand lightly touches the brim near her cheek. Her {argument name="hair style" default="black hair loosely tied in a low messy bun with wispy strands around her face"} is slightly windblown. The hat casts delicate dotted lattice shadows across her face, neck, shoulder, and upper back. She wears a {argument name="outfit" default="white off-shoulder eyelet lace summer blouse with floral cutout embroidery"}, with two thin silver swimsuit straps visible on one shoulder. Background: shallow turquoise seawater in the lower half, a softly blurred distant island on the left horizon, and exactly 3 small soft white cloud areas in the blue sky. Use bright midday sunlight, warm natural skin tones, crisp focus on the face and hat texture, shallow depth of field, realistic lens compression, airy tropical vacation mood, high-resolution editorial photography, vertical 9:16 composition.