照片AI圓夢社團咒語記錄

 115/04/08大頭娃

This is a highly detailed, stylized 3D cartoon portrait, based strictly on the uploaded image (highly accurate reproduction of human features). The head is noticeably large, while the body is slightly smaller and slender (classic cartoon proportions, head-to-body ratio approximately 1:3). Facial features are identical to the original image, but slightly exaggerated in the depiction of large, expressive eyes, sharper eyebrows, smooth, delicate skin, and natural freckles, conveying a serious and focused expression. 

Her hair is casually tied back with a few loose strands; the texture is realistic, and it's slightly tousled to enhance the stylized effect. 

She wears a white shirt with a fitted black blazer. Compared to her head, her body is more slender and compact, reinforcing the cartoonish effect. 

The image is centered, a close-up frontal shot, with a smiling face, depicting the upper body. 

The lighting is soft and cinematic, with warm tones, shallow depth of field, blurred background, and an elegant interior setting with arched decorations. 

Ultra-realistic 3D rendering, Pixar/Disney style, subsurface scattering, global illumination, 8K detail. 

Highlights exaggerated heads and faces, dramatic expressions, clean composition, and professional character design. 

Important additional features (highly effective for the generator):

Big head, small body, cartoon style 

Exaggerated proportions 

Head larger than torso 

Stylized anatomical structure, blending chibi-realistic style 

Facial similarity takes precedence over stylization 

Parameters: 

Aspect Ratio: 9:16 

Style: 3D Cartoon, Pixar Style, Comic Book 

Facial Reference: Strictly Reproduced 

Quality: Ultra-detailed, Cinematic Lighting, Masterpiece

115/04/11「我要突破」

Use only the uploaded image as a reference for the face and identity.

Do not alter the facial structure, eye shape, or proportions. Ensure the identity information is 100% accurate.

Please create a high-quality 3D animated portrait using a modern Pixar/DreamWorks style rendering.

Composition and Pose:

The figure emerges from a torn hole in a clean white surface.

The tear is circular with realistically rough edges.

The figure leans slightly forward, making eye contact with the viewer through the hole.

The upper body (above the waist) is visible in the image, centered.

Expression:

A bright, friendly smile, showing teeth.

Bright, expressive eyes, looking directly at the camera.

Energetic and cheerful.

Details (Important):

One hand holds a Canon SLR camera, facing the viewer (camera clearly visible, slightly tilted forward).

The other hand naturally grips the torn edge of the paper.

The camera strap hangs naturally.

Appearance:

Modern casual attire (pure white and blue T-shirt, minimalist design).

Hairstyle should refer to the reference image, but slightly modified for the animation.

Environment:

The background inside the hole is a smooth, vibrant solid color (preferably a blue gradient).

The exterior of the hole remains pure white, with soft shadows added around it to enhance depth.

Rendering Style:

High-level 3D animation, soft global lighting, smooth materials.

Delicate skin texture, slightly stylized (not hyper-realistic).

Rounded facial features, slightly exaggerated proportions, typical Pixar style.

Clean lighting, soft shadows, vibrant colors.

Composition:

Vertical composition, 4:5 aspect ratio.

The main subject occupies approximately 70% of the frame.

The hole is centered and clearly visible.

Quality:

Ultra-high detail, sharp focus, 8K rendering quality.

It has no extra hands and poor anatomical structure.

115/04/15膠囊公仔

This chibi-style figure is richly detailed, with a focused expression, holding a mini SLR camera to his face as if taking a picture. He wears a black baseball cap with a brown patch on the brim, a grey fitted jacket, black trousers, and black boots. A mini suitcase sits beside him, adding a touch of narrative. The figure features exaggerated chibi proportions, a large head, expressive eyes, and meticulously detailed facial features. The character is encased in a small, transparent capsule, gently held by two fingers, emphasizing proportion and realism. The capsule is crystal clear, reflecting and refracting subtle light and shadow effects. This figure is crafted with high-end collectible toy techniques, boasting ultra-realistic textures, a smooth plastic surface, and fine microscopic details. A cinematic shallow depth of field makes the subject sharp and clear, while the background is softly blurred (bokeh effect). Warm, soft, and diffused lighting creates gentle highlights and natural shadows, further enhancing the figure's premium feel. Toy aesthetics. Highly detailed 4K resolution, studio-quality rendering, realistic lighting, soft glow, professional product photography style.

"Negative Features": "Blurry, low resolution, aspect ratio distortion, cluttered background, dim lighting, unrealistic textures, noise, imperfections, oversaturated colors, extraneous limbs, incorrect human anatomy."

"Aspect Ratio": "9:16"

115/04/18突破九宮格(帶文字)

Create a set of 9 expressive 3D cartoon-style avatar stickers featuring the same character, arranged in a simple 3x3 grid layout.

The character must maintain consistent facial structure, proportions, skin tone, and identity across all images.

Style (Crucial):

Highly stylized Pixar/emoji-style 3D characters with slightly larger heads and smaller, more compact bodies (similar to chibi proportions).

Eyes must be slightly large, round, and expressive.

Facial expressions should be cute, playful, and exaggerated, rather than realistic.

Features should be soft, rounded, and possess a friendly, high-quality Pixar style.

Expressions and Variations:

Each frame must contain different expressions, different body language, and different clothing colors, including:

Winking and pointing (under Traditional Chinese text "Just like that"), cool and stylish (under Traditional Chinese text "Awesome, right?"), chin resting on hand in thought (under Traditional Chinese text "Let me think"), a calm pose with arms crossed (under Traditional Chinese text "Hmm"), giving a thumbs up (under Traditional Chinese text "Thumbs up"), drinking water (under Traditional Chinese text "Cheers"), waving (under Traditional Chinese text "Hi"), laughing heartily (under Traditional Chinese text "Haha"), and a confused/serious expression (under Traditional Chinese text "Seriously"). "Traditional Chinese characters should not be too large."

Composition (Strict):

Each head should appear to emerge from a torn hole in paper, with realistic tear edges naturally surrounding the upper body. A light gray gradient is used inside the hole.

Characters should be slightly raised (with 3D depth), not flat.

Limited to half-body (chest and above).

Grid perfectly aligned, evenly spaced, no distortion.

Lighting:

Soft, even, and clean studio lighting.

No harsh shadows. Smooth highlights.

Background (Locked):

The entire grid uses a single, cool-toned pure white background.

No gradients, no variations, no color shift between tiles.

Consistency Rules:

All 9 heads must clearly represent the same person's characteristics.

Rendering Quality:

Ultra-clear, high-resolution, sharp, and detailed 3D rendering.

Flawless, no distortion, no overly realistic tendencies, no plastic-like appearance.

Aspect Ratio:

4:5 vertical screen (perfectly adapts to the entire grid)

115/04/22  社團讓您靠

Create an ultra-realistic, high-detail portrait using the uploaded photos, keeping my face perfectly accurate with zero changes to facial structure, proportions, skin texture, or expression.

I am standing in a modern creative studio, leaning casually against a large sculptural installation spelling “照片AI圓夢社團”, arranged vertically like a bold architectural totem. Each letter is stacked upward, reaching my full height, giving the composition a strong, balanced, and powerful presence. The sculpture features a vibrant, high-gloss finish with rich color gradients flowing across the surface. Think electric blue blending into coral red, warm orange, and subtle neon yellow highlights. The letters feel sculptural, polished, and contemporary, catching studio light beautifully and creating depth and visual energy.

I lean one shoulder lightly against the vertical “zunain” structure. My body faces slightly to the side, with most of my weight resting on one leg in a relaxed yet confident stance. Both hands rest naturally inside my cargo pockets. My head tilts slightly as I look calmly toward the camera, with messy, natural hair exactly matching the reference, preserving a thoughtful, effortless expression.

I’m wearing an oversized premium t-shirt. I’ve paired it with modern cargo pants enhanced with subtle color accents such as contrast stitching_toggle tones, tonal panels, or soft reflective trims. On my feet are clean white sneakers with vivid red detailing, crisp laces, and sharply defined soles that add energy without overpowering the outfit.

The overall style is expressive modern streetwear, moving away from black-and-white minimalism and leaning into confident color, texture, and form.

The shot is captured from a low angle to amplify presence and make the vertical “zunain” sculpture feel monumental and architectural beside me.

Lighting is dynamic and studio-controlled, with soft directional light sculpting my face and outfit, complemented by colorful rim lighting and subtle ambient glow reflecting off the sculpture. Gentle color spill adds depth and atmosphere while keeping the subject clean and realistic.

The backdrop is a refined studio environment with a smooth gradient shifting from deep charcoal into rich slate blue or muted violet, maintaining focus while enhancing the colorful composition.

Captured with an 85mm portrait lens look for cinematic depth, gentle compression, and natural proportions. Ultra-sharp focus, realistic textures, editorial-grade color grading.

The final mood is bold, modern, creative, and confident, blending vibrant streetwear energy with architectural design in a striking low-angle editorial portrait.尺寸(9:16)

115/04/25 與哆啦A夢主角打彈珠

This artwork depicts a semi-realistic figure (as shown in the attached photo) rendered in a cute, 3D chibi style, playing marbles on dry ground with Doraemon, Nobita, Suneo, Shizuka, and Takeshi.

Their faces beam with joy and competition, colorful marbles scattered across the ground. The figure crouches down, intently manipulating the marbles with his fingers, his large, expressive eyes brimming with adorable charm.

The background evokes scenes from Doraemon movies: open fields on the edge of a Japanese village, simple houses, and a clear blue sky dotted with white clouds.

The ultra-detailed 3D visual style, cinematic lighting effects, depth of field, realistic land and marble textures, bright and vibrant colors, warm and cheerful atmosphere, and high-quality rendering are reminiscent of Pixar animated films. (Size: 3:4)

115/04/29 逗趣公仔

This is a highly detailed, stylized 3D character rendering depicting a lively and energetic character (male or female). It features exaggerated chibi proportions (larger head, more compact body), large, expressive eyes, smooth, delicate skin, and highly detailed, naturally flowing hair with subtle wisps.

Hair:

• Female: Long, straight or slightly tousled hair, naturally voluminous, with optional simple hair clips.

• Male: Short to medium-length stylized hair, soft texture, natural strand details.

Facial Expression: Playful and slightly defiant, eyebrows drooping, sharp gaze to one side (slanted eye), slightly wrinkled nose, tightly pursed lips, revealing a mischievous smile.

Position (Important – Must be completely identical):

Standing posture, feet together and pointing forward, arms hanging naturally at the sides, shoulders slightly forward, head slightly lowered, gaze slightly suspicious/mischievous to one side. The body remains still and compact, slightly leaning forward.

Clothing:

Loose-fitting long-sleeved sweatshirts or belted tops, made of soft, wrinkled fabric with a chest pocket; simple design, paired with simple socks and casual sneakers.

Lighting:

Soft studio lighting, soft directional key light, smooth shadows, subtle ambient light fill, and soft highlight transitions.

Rendering Style:

High-end 3D stylized realism (Pixar level), ultra-clear textures, soft subsurface scattering, and cinematic depth-of-field effects.

Background:

Clean, simple, and softly blurred warm-toned neutral background.

4:5 aspect ratio.

Vertical composition, emphasizing the height of the image, movie poster-style layout..

115/05/02 與哆啦A夢家族「樹上讀書」

The photo features casually dressed figures alongside a dressed giraffe, raccoon, Bubu, and other adorable animal-like creatures. Doraemon, Nobita, and his friends are also depicted sitting in a tree reading books, each wearing a Doraemon propeller on their head. It is presented in 3D, 3:4 aspect ratio.

115/05/06 環形展示

你是一個高端時尚視覺生成系統 + 使用者氣質分析系統 + 風格人格建模系統，

你的任務是：

基於使用者上傳的自拍，生成一張「環形空間 · 360°氣質沉浸海報」。

目標：

不是展示多個造型，而是構建一個——
👉「使用者氣質的完整空間」

讓使用者產生：
沉浸感 + 自我探索感 + 分享慾

——————————————————

一、核心邏輯（必須執行，隱藏推理）
分析使用者氣質（不要輸出過程）：

維度：

氣質：清冷 / 溫柔 / 明豔 / 鬆弛 / 知性 / 氛圍 / 少女
五官：立體度 / 柔和度 / 攻擊性 / 親和力
風格傾向：極簡 / 韓系 / 法式 / 都市 / 輕熟 / 少女
色彩適配：冷 / 暖 / 中性 / 低飽和
鎖定「主氣質區間」
在該區間生成 5–7 個「氣質分身」（風格分支）

要求：
✔ 全部適合使用者
✔ 風格連續變化（有梯度）
✔ 不跳躍、不割裂

——————————————————

二、核心視覺結構（關鍵差異點）

👉 構建一個「環形空間（circular environment）」
而不是平面排版

空間設定：
一個抽象、極簡、連續的空間（例如：柔和曲面空間 / 無邊界室內空間，禁止用例如選項）
地面 / 背景 / 光線是連續的（非常關鍵）
形成「360°包圍感」
人物結構：

在這個空間中：

👉 同一個人（同一張臉）以不同風格
👉 出現在空間的不同位置（像分身）

分布方式：

沿環形路徑分布（半圓或整環）
彼此之間有距離，但屬於同一空間
中心角色：
中間放一個「主版本」（最接近真實）
或輕微強化後的使用者形象，比例不用很大，主要是為了凸顯分身氣質。

👉 作為視覺錨點

——————————————————

三、人物生成要求（最關鍵）
必須強約束：

same person, same face, same identity,
no face change, no different person,
consistent facial features across all versions

每個分身變化維度：
髮型（直 / 捲 / 綁髮 / 慵懶）
妝容（清冷 / 裸感 / 氛圍 / 精緻）
穿搭（極簡 / 溫柔 / 鬆弛 / 都市 / 輕熟）
姿態（站立 / 回頭 / 側身 / 端坐）
風格示例（自動生成）：

清冷知性 / 溫柔氛圍 / 鬆弛韓系 / 都市極簡 / 輕熟氣場 / 日常休閒 / 少女活潑

選擇5-6個風格

強調：

👉 必須像真實模特拍攝
👉 禁止cosplay感
👉 禁止誇張設計

四、空間與光影（高級感來源）
空間關鍵詞：

panoramic environment,
circular space,
immersive scene,
continuous background,
soft curved architecture

光影設計：
整體統一光源（避免割裂）
局部有變化（強調不同風格）

例如：

左側偏冷光（清冷）
右側偏暖光（溫柔）
中間自然光（真實）

👉 形成「氣質流動」

氛圍：
輕體積光（volumetric light）
空氣感 / 輕霧（subtle haze）
柔和陰影

——————————————————

五、資訊層（弱UI，避免破壞沉浸）

不要做強卡片UI，而是：
每個分身（疊加在遮罩上）
遮罩：
bottom gradient overlay
dark to transparent
opacity 60–80%
cinematic fade

內容：
風格名稱（大字）
一句話標籤（有情緒張力）
穿搭關鍵詞（3個）
色彩點（2–3個）


頂部：

主標題：

👉 你不只有一種樣子

副標題：

👉 Explore your style spectrum

底部（互動）：

👉 你最想成為哪一個？
👉 留言區告訴我

六、構圖方式

優先使用：

環形構圖（circular composition）
S型視覺流動（S-curve flow）
視線引導圍繞中心旋轉

七、色彩系統

整體：

低飽和
黑 / 灰 / 米白為基底

局部：

霧藍 / 燕麥 / 豆沙 / 奶油色

👉 不同分身可偏色，但整體統一

八、品質要求

editorial fashion photography,
high-end magazine style,
realistic skin texture,
soft lighting,
8k detail,
clean composition

九、禁止項

❌ 多人臉（必須同一人）
❌ 風格斷層
❌ 服裝混亂
❌ 網紅濾鏡
❌ UI過重
❌ 電商感

十、最終輸出

生成一張：

👉 9:16 直式
👉 環形空間沉浸視覺
👉 同一人物多氣質分身
👉 高級感 + 連續空間 + 強代入

115/05/09 環遊世界

This is a vertical travel magazine-style collage (9:16 aspect ratio) depicting a journey to Lake Toba, Indonesia, in a modern and vibrant style. The image utilizes overlapping elements, with torn paper textures at the top and bottom. A striking white title, "Around the World," is positioned at the top. At the center of the collage is a full-body photograph of a young tourist (reference photo: denim shirt, black T-shirt, backpack, khaki pants), holding a map and digital camera. A white outline highlights his identity. Around him, the collage is filled with iconic views of Taipei 101: landmarks of Kyoto, Japan; beautiful lake and mountain scenery; the San Francisco Bridge; ferries; and two dancers in traditional Taiwanese aboriginal costumes. At the very bottom, above a background of mountains and a suspension bridge (cable car), a line of small print reads "照片AI圓夢社團" The entire composition uses a retro-style paper-textured background with a polka-dot pattern (half-tone).

115/05/13 辦公室娃娃

Please create a chibi-style doll based on the attached image, striving for 100% accuracy in recreating the person in the photo. This is a richly detailed chibi-style doll, an adorable BJD figure, small and exquisite, with a relatively large head, realistic skin tone, bright eyes, fluffy black hair, light makeup, and small lips. The doll wears a loose black sweatshirt printed with the words "照片AI圓夢社團" holding an iced coffee and bubble tea. The mini drinks are accompanied by realistic ice cubes and pearls, and the doll is sipping the bubble tea with a straw. The background includes a desk, computer monitor, keyboard, and a blurred colleague. The doll's realistic hands support its own feet, which are clad in gray trousers. The design utilizes shallow depth of field, soft natural lighting, cinematic photography, high detail, realistic toy photography, cute aesthetics, macro lens, realistic texture, and is a high-end collectible doll, creating a warm office atmosphere. The size is 3:4.

115/05/20 情侶照

Ultra-detailed stylized 3D character render of a playful young couple (male and female), designed with exaggerated chibi proportions (oversized heads, compact bodies), large expressive eyes, soft smooth skin, and highly detailed hair strands with natural color, natural flow and subtle flyaways.
Characters:
Female: long straight or slightly tousled hair, natural volume, optional minimal hair clip
Male: short to medium stylized hair, soft texture with clean strand detail, and grooming details
Facial Expression (IMPORTANT):
Both characters have a mischievous, slightly defiant expression
Eyebrows slightly lowered
Eyes looking sharply to the side (side-eye)
Nose subtly scrunched
Lips pressed into a tight playful smirk
Pose (IMPORTANT — match exactly for BOTH):
Standing back-to-back, close together
Feet close and pointing forward
Arms relaxed straight down at their sides
Eyes looking sideways (same direction or slightly toward each other)
compact posture
Interaction Detail:
Their backs lightly touching, giving a subtle cute couple vibe, but both maintain the same pose (no hugging or hand-holding)
Clothing:
Oversized long-sleeve cream beige sweatshirts, and maching trousers (minimal design, soft folds, chest pocket)
Simple socks and casual sneakers
Slight color contrast between male and female outfits
Lighting:
Soft studio lighting
Gentle directional key light
Smooth soft shadows
Subtle ambient fill
Soft highlight transitions
Rendering Style:
High-end 3D stylized realism (Pixar-like quality)
Ultra-clean textures
Soft subsurface scattering
Cinematic depth of field
Background:
Clean, soft bokeh seamless studio backdrop with warm tones. No objects in the background.
Composition:
4:5 aspect ratio
Vertical portrait
Full body, centered
Cinematic poster layout
Balanced spacing between characters.

115/05/24 深情照片

Create a stylized big-head miniature portrait in a 9:16 aspect ratio. Two persons appear as tiny characters with slightly enlarged heads and small bodies, sitting together inside a single large human hand.

Pose: Both are seated close together, cross-legged on the palm. The girl has a pouty, angry expression, looking slightly away from the boy. The boy has a soft, gentle smile and is leaning slightly toward her, trying to calm her, looking at her with a caring expression.

Interaction: The boy’s body language should feel soft and apologetic. The girl looks cute but clearly upset.

Environment: Blurred city night background with colorful bokeh lights.

Outfits: Both are wearing modern casual streetwear outfits. The girl wears a simple stylish casual outfit (hoodie or cropped jacket with jeans or relaxed pants). The boy wears a clean casual outfit (plain t-shirt or hoodie with jeans or cargo pants). No traditional clothing.

Lighting: Soft cinematic lighting, shallow depth of field, ultra-realistic, high detail, clean focus.

Face lock: Do not change faces, 100% identity preservation.