YouMind
登錄

Sketch / 線條畫

在 YouMind AI 提示詞庫裡探索免費的 Sketch / 線條畫 圖片 提示詞,歸類於 風格。每一條都精選自真實的創作流程,可以直接複製、改用與重用。

作者 @Sairah
GPT Image 2
復古故事書風格旅行插畫提示詞
查看完整提示詞

將上載的照片轉化為精美的手繪編輯風格旅行插畫,呈現迷人的復古故事書美學。 風格:在質感溫暖的紙張上,以傳統色鉛筆及蠟筆繪製的插畫。可見素描筆觸、顆粒感顏料、不完美的手繪輪廓、細膩的交叉排線,以及柔和的啞光效果。融合歐洲旅行手札藝術與現代插畫敘事風格。 主體:保留原圖人物的辨識特徵、五官、髮型、服飾、姿勢及重要細節。運用優雅簡練的造型、富表現力的神態、細緻的線條、紅潤的臉頰,以及自然的比例,營造略帶奇幻色彩的手繪觀感。 背景:按需要以畫意般的建築細節、優美風景、迷人建築物、藍天、連綿山丘或鵝卵石街道,重現原有環境。在保持景深與透視的同時,將細節簡化為富表現力的手繪形態。 配色:暖米白、赤陶紅、灰藍色、柔和橄欖綠、炭黑色及淺米色。採用和諧的色彩、柔和的對比,以及微妙的復古色調變化。 質感與細節:在整幅圖像中加入清晰可見的鉛筆刮痕、粗糙的蠟筆筆跡、不均勻的上色、精緻的排線,以及自然的紙張紋理。讓作品呈現觸感豐富、自然有機且真正手工繪製的效果。可視情況加入細小的手繪心形或俏皮點綴。 構圖:3:4 垂直人像比例,構圖平衡,滿版插畫配以略微不規則的紙張邊緣。確保主體突出,背景引人入勝而不喧賓奪主。 最終效果:一幅精緻而充滿懷舊感的旅行手札插畫,猶如由專業編輯插畫師親手繪製。避免照片般的寫實效果、光滑的數碼渲染、3D 效果、過度平滑的處理,以及千篇一律的動漫風格。

產生圖片
作者 @小小东
GPT Image 2
鯨魚尾鰭攝影與墨繪插畫雙聯作品
查看完整提示詞

創作一幅垂直雙聯畫風格的作品,畫面分為上下兩個截然不同的部分。上半部分是一張照片級真實的高解像度相片,展示座頭鯨的尾鰭從深藍色的海洋中垂直躍出水面。尾鰭底部周圍的海水泛起白色泡沫與飛濺的水花,質感分明。背景是柔和對焦的藍色山脈,襯托着明亮的天空與散落的雲朵。下半部分則是極簡風格的插畫,以淡藍灰色的紙質紋理作為背景。畫面呈現同一鯨魚尾鰭形狀的黑色墨水風格化繪圖,但完全採用點描或點刻技法(由數以千計的細小圓點組成)繪製。請在下半部分加入特定的韓文排版:左上角以小字標示「No. 027」,接着是大標題「먼바다로 가는 꼬리」,以及較小的副標題「먼 곳을 향한 꼬리」。在畫面最底部的正中央,以襯線字體標示學名「Megaptera novaeangliae · 해양 포유류」。

產生圖片
作者 @Ciri
GPT Image 2.5
極簡紙質質感藝術封面
查看完整提示詞

極簡手繪紙本插畫 將上傳的照片轉化為極簡風格、帶有紙張質感的手繪藝術封面。 構圖 - 主體只佔畫面約 10–20%。 - 將主體置於構圖底部中央附近。 - 在整幅圖像中保留大量留白。 - 僅以寥寥數筆或細小色塊暗示環境。 保留 - 原主體最易辨識的特徵。 - 基本的輪廓與比例。 - 原本的姿態或主要動作。 - 重要物件與標誌性細節。 - 核心故事、人物關係及情感氛圍。 色彩 - 從原照片中提取主色調。 - 將用色限制於四種主色以內。 - 使用大膽而節制的平塗色塊。 - 保持和諧柔和的色彩搭配。 - 加入細緻的紙張紋理與手繪筆觸質感。 簡化 去除多餘細節與視覺雜訊,只保留令主體一眼可辨的必要視覺資訊。 字體排版 - 可選擇加入少量簡單文字。 - 文字須精簡含蓄,兼具美感。 - 讓文字自然融入構圖之中。 - 營造藝術書籍封面、獨立出版、現代編輯設計或靜思繪本的視覺語言。 - 若文字無助提升構圖,切勿勉強加入。 風格與媒材 - 細膩且略帶不完美的手繪線條。 - 少數輪廓分明的壓克力平塗色塊。 - 粗糙而富觸感的紙張質地。 - 清晰可見的手繪筆觸。 - 有機、不規則的自然輪廓。 - 微妙的瑕疵感,令作品散發真正的手作氣息。 整體氛圍 寧靜、詩意、雅致、極簡、純真、悠然、富藝術感、耐人尋味、別樹一幟且具高級感。 避免 - 彩色鉛筆渲染效果。 - 乾筆刷效果。 - 水彩暈染或滲透效果。 - 純線稿插畫。 - 過於精細或寫實的插畫。 - 厚重的厚塗或油畫質感。 - 光滑精緻的數碼插畫。 - 3D 渲染、光澤表面或人工燈光效果。 - 千篇一律的商業卡通風格。 - 電商廣告美學。 - 通用海報模板。 - 過多的裝飾元素。 - 過於複雜的構圖。 - 過多文字。 最終效果:一幅精緻、極簡的手作紙本插畫,留有充裕空間、用色克制、質感細膩,並散發出獨立藝術書籍封面的高雅氣質。

產生圖片
GPT Image 2.5
旋轉木馬照片轉刮板藝術(分屏效果)
查看完整提示詞

垂直分割畫面構圖。上半部分是一張寫實照片,拍攝一對年輕情侶坐在旋轉木馬上,沐浴在四周燈泡散發的溫暖金光之中。一名留着深色捲髮、身穿深藍色襯衫的男子,抬頭望向一位長棕色波浪捲髮、穿着紅白條紋毛衣的女子;兩人雙手握着黃銅扶手桿,含情脈脈地相視而笑。下半部分將同一場景轉化為象牙色紙張背景上的高對比黑白刮畫插圖。此部分的圖像以負片風格繪製,運用大膽的墨塊與白色刮線勾勒出人物的面容和衣飾。插圖四周點綴着手寫風格的文字元素:左上角為「CAROUSEL DAYS」及「laugh. spin. remember.」,右上角為座標「lat. 35.7° N / long. 139.7° E」,左下角則為「CAROUSEL DIARY.」與「a little world that moves in circles.」。

產生圖片
作者 @Harboris
GPT Image 2.5
Scratchboard diary art poster
查看完整提示詞

Create a high-resolution, vertical 2:3 editorial art poster using the uploaded reference photo. TOP HALF: Use the original photograph full bleed, preserving the subject's identity, appearance, natural lighting, and photographic quality. Crop carefully if needed. No borders or margins. BOTTOM HALF: On textured ivory paper, create a monochrome scratchboard illustration of only the main subject and essential accessories. Place a large, rough-edged black ink block in the center. The figure must be larger than the block, with roughly two-thirds inside it and the lower third breaking out onto the ivory paper. Add loose horizontal ink strokes beneath the feet for grounding. Inside the black block, use fine white and light-grey etched linework. Outside it, continue the same figure in expressive black pen strokes on ivory. Ensure a seamless transition between both tonal styles. Avoid excessive hatching and unnecessary background details. COLORS: The bottom half must use only black, white, and ivory, with subtle paper grain. TYPOGRAPHY: Use rough, upright handwritten lettering. - Top left: a 2 to 4 word title across two lines, followed by three short lowercase verbs separated by periods. - Top right: latitude and longitude on two lines only. - Bottom left: the location or theme followed by “DIARY.”, with a short descriptive sentence beneath. - Keep all typography black on ivory, natural and imperfect, never cursive. FINAL COMPOSITION: Exactly one standalone poster with two equal horizontal halves and one perfectly straight dividing line. Preserve the original subject's identity and proportions. Maintain generous negative space and a restrained, collectible editorial aesthetic, combining documentary photography with expressive hand-printed illustration. Print quality, 2:3 vertical aspect ratio.

產生圖片
作者 @Sairah
GPT Image 2
Hand-Drawn Storybook Illustration
查看完整提示詞

Transform the uploaded photo into a beautiful, heartwarming, hand-drawn storybook illustration. Preserve the original subjects’ recognizable features, hairstyles, poses, expressions, clothing, body language, and emotional connection. Keep the composition and interaction faithful to the reference photo. ART STYLE: Warm, playful, modern children's book illustration with a soft hand-painted, colored-pencil, and crayon texture. Use expressive black-brown sketch outlines, slightly imperfect lines, simplified organic shapes, gentle facial features, rosy cheeks, and natural, joyful expressions. Apply a warm pastel palette with coral pink, dusty blue, caramel brown, soft peach, and creamy off-white. TEXTURE & DETAILS: Visible pencil strokes, subtle paper grain, lightly uneven coloring, sketchy hair strands, minimal shading, and charming handmade imperfections. Create a cozy, nostalgic, affectionate atmosphere full of warmth and personality. BACKGROUND: Simplify the original background into a clean, minimal cream-colored paper backdrop. Add a few small, hand-drawn decorative marks, playful lines, tiny hearts, or expressive doodles that complement the scene without distracting from the subjects. COMPOSITION: Maintain the original framing, subject placement, pose, and interaction. Make the subjects the clear focal point. Use clean shapes, balanced negative space, and an organic editorial storybook aesthetic. IMPORTANT: Illustrate the actual uploaded photo rather than inventing a new scene. Preserve the subjects' identities and defining characteristics. Avoid photorealism, 3D rendering, glossy digital painting, excessive detail, harsh shadows, complex backgrounds, and overly polished vector graphics. No text, watermark, or border. OUTPUT: A high-quality, charming, hand-drawn storybook illustration with a soft vintage print feel, warm colors, visible artistic texture, and genuine emotional expression. Adapt the illustration naturally to the uploaded photo, regardless of the subjects or setting.

產生圖片
作者 @小小东
GPT Image 2
GPT Image 2 Hand-drawn Infographic Prompt
查看完整提示詞

Two vertically connected rounded rectangular panels are anchored by extremely thick, slightly undulating black hand-drawn outlines, with white buckles crossing the panel edges, forming striking paired modules and clear vertical rhythm. Content is organized using Gestalt proximity and layout zoning, with each panel internally divided into a stable left-image/right-text column, balancing left and right weights, titles and information read top-to-bottom in the right column; the overall visual center falls on the junction of the two panels and their respective prominent large titles, keeping the page header area independent with logos and main title. Images and text connect via juxtaposition and backing: the left side is reserved for character or theme illustrations, placed on irregular rounded clean light-colored bases; the right side uses prominent large characters to lead multiple lines of explanation, accompanied by small line-drawing symbols adjacent to text to indicate content. Text hierarchy is always distinct, large titles are thick and striking, explanatory text is thinner with generous line spacing, small text and symbols create rhythm points; glyph construction is upright and readable, with clear counters, crisp endings, and compact faces complementing handwritten images, maintaining this hierarchy regardless of text volume; when text is insufficient, secondary small text is supplemented based on the theme. Text edges are flat and clear, forming a clean typography layer outside the hand-drawn lines. Whitespace includes gaps between panels, backing whitespace around image bases, reading gaps between text lines, and pauses between header and body; each remains spacious, allowing thick frames, illustrations, and titles to stand out clearly. Illustration content is provided by the appended theme, adopting a complete, concentrated, slightly tilted hand-drawn composition within the left column, with simple main outlines, light dynamic details, and upright load-bearing parts; content density is about half-frame, mainly concentrated in the left column of each panel, the rest being white or light-colored rounded bases, flat panel color blocks, and bottom whitespace, generating no scenery. Uses a composite of techniques: outlining frames and characters with continuous, slightly shaky thick black lines; spreading large panel areas and local illustration colors with clean, smooth digital flat coloring; pressing information areas with crisp text typesetting; overall resembling an illustration arrangement synthesized from hand-drawn lines, computer flat coloring, and clear typesetting. Color scheme maintains a high-brightness keynote, colors vivid, saturated but not glaring, the most intense color surfaces fall within the paired panels echoed by character locals; pairing warm light backgrounds with varying brightness levels of light color surfaces, stabilizing layers with strong black-and-white line text contrast. Overall warm and cheerful, friendly and energetic, resembling matte digital color blocks combined with dense marker lines; hue follows the theme, but brightness/darkness, thickness/thinness, saturation/dilution, warm tendencies, and cheerful friendly psychological feelings maintain this personality regardless of theme. Overall temperament is light, friendly, and orderly. Appended theme, accurate text, usage, and aspect ratio take priority; grouping of upper/lower panels, left-image/right-text, thick black hand-drawn borders, image backing, clear text hierarchy, and high-brightness warm-thick color schemes remain unchanged. ------ Theme: Water Cycle Educational Poster

產生圖片
作者 @Arisha Ai
GPT Image 2.5
Vintage Cowboy Portrait
查看完整提示詞

Create an ultra-detailed artistic portrait of a beautiful young woman wearing a vintage cowboy hat with decorative dark stitching and round, thin-framed eyeglasses. She has long, flowing dark-brown hair with subtle caramel highlights, expressive brown eyes, naturally full lips, and a gentle smile. She is wearing a cream-colored cable-knit sweater with a small embroidered polo player logo on the chest and a leather strap across her shoulder. The image should have a mixed-media aesthetic, blending photorealism with charcoal sketching and ink splatters on a textured beige paper background.

產生圖片
作者 @K@miya.
GPT Image 2.5
Professionalize Rough Storyboard Sketches
查看完整提示詞

Using the provided reference image, transform the rough stick-figure storyboard into a highly detailed, professional manga-style storyboard. Keep the original 6-panel layout and grid structure intact. **Visual Style:** - Convert all crude stick figures into polished, dynamic anime/manga character designs with proper anatomy and clothing (school uniforms, sports gear). - Apply heavy cross-hatching, speed lines, and dramatic shading to create a high-energy action atmosphere. - Maintain the black-and-white ink aesthetic typical of Japanese manga storyboards. **Panel-by-Panel Instructions:** 1. **Panel 1:** Depict a broadcasting club member holding a microphone, shouting enthusiastically. Add background elements suggesting a school setting. 2. **Panel 2:** Show a chaotic crowd of students from various clubs running forward in a stampede. Use motion blur and intense expressions. 3. **Panel 3:** Illustrate a soccer player kicking a ball with force. The ball should have 'burning' or energy effects trailing behind it. 4. **Panel 4:** A baseball player looking up in shock as the glowing soccer ball flies directly toward their face. 5. **Panel 5:** A brass band member playing a tuba/sousaphone aggressively, creating visible sound waves that knock the baseball player down. 6. **Panel 6:** A dramatic close-up of a Kendo club member wearing a mask (men) and hakama, posing heroically with a bamboo sword raised against a bright sunburst background. **Text & Layout:** - Preserve the Japanese text columns on the right side (Visual Image/Audio Memo), but refine the handwriting to look like neat storyboard notes. - Ensure the header includes the title "部活動対抗棒倒し" (Club Activity Stick Pushing Competition).

產生圖片
作者 @Shore Lyn
GPT Image 2.5
Before-After Doodle Background Photo Edit
查看完整提示詞

Create a vertical 3:4 before-and-after image split into two stacked panels of equal height. TOP PANEL (before): the original photograph exactly as provided, completely unedited — same subject, pose, background, colors and lighting. Full bleed to the panel edges. BOTTOM PANEL (after): the same person or people, still fully photographic and realistic, cut out from the original photo and placed centered on a warm cream off-white paper background. Keep their face, identity, expression, pose, clothing and proportions exactly as in the photo — never illustrate, stylize, smooth or redraw the subject. Clean cutout edges including hair. Replace the entire background with charming hand-drawn doodles in thin black pen line art — monoline, no fill, light and loose like a sketchbook, with a little character and personality in the linework. Use about eight doodle elements, balanced across the panel. Draw one or two simplified landmarks or buildings from the original background, one on each side of the subject at small scale. Add two or three small cartoon figures doing light actions that suit the scene — riding a bicycle, walking a dog, taking a photo, waving from a shop, holding a balloon. Add three or four small accents scattered in the open areas: a cloud, a sparkle or two, a paper plane with a dotted trail, a musical note, a few flowers or small plants near the ground. Compose it with breathing room. Keep every doodle at roughly a third of the subject's height or less, and rest the buildings and figures along a shared ground line so the panel has a clear horizon. Let the accents float higher in the empty upper corners to fill the space lightly. Leave a clear margin of untouched cream around the subject on all sides — doodles never overlap or touch them — and keep visible cream gaps between neighbouring elements so nothing clusters. Ground the subject with a thin sketchy horizontal line under the feet with a few short hatch marks and a soft natural shadow. Keep it light, not a dense scribble. Place a short handwritten caption in the open upper-left area: three lines, three to five words each, casual lowercase handwriting in thin black ink, warm and reflective in tone. Add one small red hand-drawn heart just below it with space around it. Everything in the lower panel is black line on cream, except three or four tiny color accents — a red heart, pink flowers, a yellow sparkle. The subject alone stays in full photographic color. Avoid: a cluttered or busy background, doodles filling every corner, doodles overlapping or touching the subject, large doodles rivaling the subject's size, three or more buildings, dense scribbled shadows, heavy thick linework, shading or crosshatching, long caption text, more than three lines of text, an illustrated or cartoon subject, changed face or identity, beautified skin, altered clothing or pose, colored or filled doodles, 3D rendering, side-by-side panels, watermark.

產生圖片
GPT Image 2.5
GPT Image 2.5 Manga App Review
查看完整提示詞

A vertical manga page in black and white featuring three characters: an energetic boy with spiky light hair, a calm boy with dark messy hair, and a girl with long light hair. The layout consists of five panels arranged vertically. Panel 1 (Top): Close-up of the energetic boy raising his fist triumphantly with speed lines behind him. He shouts "I got an app made for GPT Image 2.5!!". To his right, the girl and the dark-haired boy look on with neutral or slightly confused expressions. Panel 2: The energetic boy is smiling happily on the left. In the center, there are three floating polaroid-style images showing examples of AI art: a landscape, a portrait of a girl, and a fluffy dog. On the right, the dark-haired boy points a finger up and explains, "With this, you can output for about 16 yen per image. It's easy to understand when converted to Japanese yen." Panel 3: The dark-haired boy looks skeptical/sweating on the left, thinking "The main point is that it runs directly without other services...". On the right, the energetic boy smiles broadly with closed eyes, saying "This manga was also output from the app!" Panel 4: The dark-haired boy is on the left, looking impressed/smiling with a sweat drop, saying "Just by writing a simple description of the desired function, I could ask for it and get it done." On the right, the energetic boy gives a thumbs up and says "The code for making the app is amazing too!" Panel 5 (Bottom): The two boys are on the left, laughing heartily with their hands raised (labeled with sound effects 'Wai'). On the right, the girl holds up a spiral notebook with a deadpan expression. The text on the notebook reads: "This manga has no punchline..."

產生圖片
作者 @Harboris
GPT Image 2.5
Photographic to Ink Wash Diptych Prompt
查看完整提示詞

Vertical 3:4 canvas split into two strictly equal horizontal halves: top 50% and bottom 50%. The upper half stays photographic; the lower half becomes the continued Second World. Together they must read as one scene. the UPPER HALF Keep the photo faithful. Preserve the subject, pose, spatial relationships, color atmosphere, and natural light. Do not redesign, repaint, replace, or restage it. Only necessary proportional cropping is allowed. CONTINUITY (highest priority) Identify one source structure that naturally reaches the center split: a road, shoreline, water, reflection, branch, beam of light, piece of architecture, or body movement. This structure must cross the boundary and continue into the lower half. The Second World begins from the photograph itself, so the same scene passes through the seam and simply changes its physical rules. TRANSITION EDGE The center may use an irregular torn-paper edge, but the tear must follow the source structure, never act as decoration. Source elements may touch, follow, break through, or extend beyond edge. Never force the same tear shape onto every image. The Second World LOWER HALF Ivory paper with subtle fibers and abundant negative space. Continue the chosen structure downward from the exact point where it meets the split, then reinterpret it with restrained photo fragments, cut-paper forms, and minimal black hand-drawn lines. The lower world must stay visibly attached Never isolate it as a separate portal, window, stage, platform, or floating vignette unless it clearly comes from the photo. SECOND WORLD LOGIC Ask: if this structure became touchable, usable, enterable, or changeable, what would it naturally become? Create one image-specific interaction. For example, water may stay attached while being pulled, a road may continue as a drawn path, or light may become something held. Do not repeat actions mechanically. FIGURES Add 0-3 tiny black line figures only when useful. They must physically interact with the continued structure. If the photo already contains strong human action, add none. Never use figures as decoration. World CAPTION One short handwritten English caption (2-6 words) based on the action. Natural, light, slightly witty. Spell it correctly. No inspirational quotes and no fixed "Same..., different..." phrasing. STYLE Real photography above, warm paper below, minimal black line doodles, subtle handmade collage texture, independent-magazine mood. Bright, airy, restrained. It should feel as if the Second was already hidden inside the photo and simply continued downward. NEGATIVE No detached lower-half illustration, no generic torn-paper template, no isolated portal, no unrelated vignette, no broken continuity, no dense illustration, no random decoration, no crowd, altered photo colors, no redesign of the upper image, no glossy 3D, no scrapbook clutter, no gibberish text, no watermark, no logo, no UI.

產生圖片
GPT Image 2.5
Architectural Art Poster Prompt
查看完整提示詞

Create a premium 2:3 vertical architectural art poster of [BUILDING / LANDMARK], combining precise architectural sketching with expressive ink and watercolor. COMPOSITION: Feature one architecturally accurate building in a dynamic three-quarter perspective, occupying 60–70% of the composition. Preserve its recognizable silhouette, proportions, structural details, and authentic architectural character. ART STYLE: Intricate hand-drawn ink lines, graphite sketching, delicate crosshatching, translucent watercolor washes, organic imperfections, and expressive ink splashes concentrated around the building’s base. TECHNICAL DETAILS: Integrate fine cyan and muted-teal construction lines, perspective guides, measurement arrows, geometric diagrams, and a few subtle handwritten annotations that relate to the building’s actual structure. COLOR & BACKGROUND: Pure white background, charcoal and graphite linework, pale cyan drafting marks, muted blue-gray watercolor, and restrained indigo accents. Keep the composition airy with generous negative space. TYPOGRAPHY: Add a small, elegant museum-archive label containing the building name, location, architectural style, and a short, factual 2–3-line English curatorial note. Keep typography refined and secondary. FINAL AESTHETIC: A collectible museum-quality architectural study blending technical precision, expressive handmade artistry, and sophisticated editorial design. AVOID: Photorealism, 3D rendering, vector-perfect outlines, inaccurate architecture, unrelated buildings, clutter, colored backgrounds, excessive annotations, logos, and watermarks. FORMAT: Full-bleed 2:3 portrait, high resolution, exceptionally detailed, authentic ink-and-watercolor craftsmanship. Generate for new york

產生圖片
GPT Image 2.5
Manga Girls Pushing Each Other Away
查看完整提示詞

A black and white manga-style illustration of three young women standing in a row against a plain white background. The central figure has light hair, animal ears (kemonomimi), and an angry expression with furrowed brows; she wears a loose t-shirt and jeans, extending both arms to push the other two girls away. The girl on the left has dark hair tied back, glasses, and a surprised open-mouthed expression, wearing a sweater and tight pants while being pushed by her shoulder. The girl on the right has messy dark hair, an annoyed shouting expression, and wears an oversized long-sleeve shirt with baggy trousers, also being pushed away. The art style features clean line work, screentone shading, and dynamic character poses typical of shonen or seinen manga.

產生圖片
正在載入更多提示詞

常見問題

YouMind 的提示詞從哪裡來?
YouMind 會從創作者社群和公開創作流程中精選高信號 AI 提示詞案例。這個庫優先收錄值得查看、複用和改寫的提示詞。
圖片提示詞分類是怎麼劃分的?
圖片提示詞按使用場景、風格和主體劃分。使用場景描述要完成的任務,風格描述視覺處理方式,主體描述畫面主要對象。
提示詞是怎麼被歸到分類裡的?
YouMind 會用 AI 分析每條圖片提示詞的正文、標題、元資料,以及可用的生成結果,然後匹配最接近的類目。隨著類目體系演進,編輯也可以繼續校正。
YouMind 是什麼?
YouMind 是一個 AI 工作空間,幫助用戶把研究、想法和創作參考轉成可用產出。提示詞庫是其中公開的一部分,用來幫助創作者更快學習優秀案例。