Photorealistic Candid Street Portrait with Extreme Shallow DOF

A highly detailed prompt for generating a photorealistic, candid street-fashion portrait. The core intent is to achieve extreme shallow depth of field, keeping a single subject tack-sharp in the mid-ground while rendering a moving crowd and the city backg

範例提示詞
Core Intent: Create a photorealistic candid street-fashion portrait with extreme shallow depth of field where a single woman is sharply in focus in the mid-ground, surrounded by a moving crowd rendered as soft blur/bokeh. Preserve the exact framing, subject placement, color grade, and atmosphere. 1) Composition & Framing (LOCK) •Aspect: 4:5 vertical. •Camera height: chest-to-eye level (slightly above the subject’s shoulder line). •Framing: mid-ground subject centered slightly left-of-center, from upper chest to top of head clearly visible; shoulders included; background heavily blurred. •Foreground: two large blurred human shapes frame the subject: •Left foreground: blurred middle-aged man occupying left third, grey/blue jacket, face out of focus. •Right foreground: blurred person occupying right-lower quadrant (dark hair), also out of focus. •Mid-right: blurred older bald man head/shoulder visible behind subject’s right shoulder. •Background: dense city street with yellow taxi/cab shapes and warm orange bokeh lights; tall urban forms dissolve into creamy blur. 2) Subject (Primary) •Person: young adult woman, fair/light skin, subtle freckles, natural editorial beauty. •Hair: medium brown, messy low updo/bun with loose wisps; slightly damp/urban texture; flyaways framing face. •Expression: calm, slightly serious/curious; lips relaxed; eyes looking directly toward camera with a soft, neutral intensity. •Pose: body facing right (3/4 profile), head turned back over left shoulder to look at camera. 3) Wardrobe & Accessories (Exact Feel) •Outerwear: {argument name="outerwear color" default="beige/tan"} trench coat (matte cotton), collar open, classic lapels, slightly rumpled city-wear texture. •Bag: small shoulder bag with thin gold chain strap on her right shoulder; strap catches a bit of light; bag itself mostly hidden. •Jewelry: small earring visible (subtle, warm tone). 4) Environment (Street Mood) •Busy metropolitan sidewalk/street scene, crowd density high. •Background palette: teal/green-grey city haze + yellow cab + warm orange bokeh lights. •No readable signage, no sharp background details—everything behind subject remains creamy blur. 5) Lighting (LOCK) •Natural daylight, overcast-to-soft sun (diffused), no harsh shadows. •Subject face evenly lit with gentle highlights on cheekbones and nose; cinematic soft contrast. 6) Camera & Optics (CRITICAL) •Look: telephoto candid portrait with strong compression. •Lens: 85mm–135mm range, wide open (f/1.4–f/2.0). •Focus: tack-sharp on subject’s eyes and eyelashes; foreground and background strongly out of focus. •Depth: creamy bokeh, smooth falloff, mild cat-eye bokeh at edges. •Shutter/feel: background crowd appears naturally blurred from DOF; slight motion blur suggested in foreground figures. 7) Color Science / Grade (LOCK) •Cinematic teal/green wash in shadows and background. •Warm highlights (skin tones natural, slightly warm). •Yellow taxi pops softly; orange bokeh lights glow. •Fine
AI生成
AI模型
Nano-banana-pro
作者

常見問題

關於character style提示詞,你想知道的一切

如何使用這個character style提示詞?

只需點擊上面的「AI生成」按鈕,即可跳轉到我們的AI圖像生成器,並且這個Character Style提示詞已自動填好。你可以進一步自訂或直接生成圖像。我們的AI會解讀提示詞,並根據你的描述創建出令人驚嘆的character視覺效果。

我可以根據特定需求修改這個提示詞嗎?

當然可以!隨意複製並定製這個提示詞,以實現你的創作構想。對於character藝術作品,嘗試添加風格、光線、構圖或情緒等細節。你的描述越具體,AI就越能理解你的創意。瀏覽我們的character提示詞合集,學習更多技巧。

哪些AI模型最適合生成character style圖像?

這個character提示詞適用於大多數現代AI圖像生成器,包括Nano Banana、Flux和GPT-Image。不同的模型各有專長——Nano Banana擅長character風格,而Flux則能生成逼真的照片級效果。

怎樣才算是一個好的character style提示詞?

優秀的character提示詞通常包含:具體的主題細節、藝術風格參考、光線條件(如柔光、戲劇性陰影)、色彩搭配、情緒/氛圍以及技術規格(如4K、細節豐富、高清)。研究我們的範例提示詞,可以學到有效的模式。

這個提示詞可以免費使用嗎?

是的!我們庫中的所有提示詞,無論是個人還是商業用途,都完全免費。你可以隨意複製、修改,並創作出色的character藝術作品,沒有限制。

More Prompts ➜
HUD battle power overlay for any character image - AI Image Prompt
HUD battle power overlay for any character image
A two-part setup: a first line to set the HUD color and a core task prompt that overlays a full-screen scouter-style analysis HUD with a battle power value chosen by Nano Banana Pro from the image’s impression.
by てんねん
AIFocus Twitter banner with Google branding - AI Image Prompt
AIFocus Twitter banner with Google branding
A Chinese prompt to create a Banana-themed Twitter banner for AIFocus that matches Google’s brand tone and design elements, suitable for social media branding.
by AI焦点
Chess legends hanging out in a bar - AI Image Prompt
Chess legends hanging out in a bar
A simple scene prompt asking for Bobby Fischer, Mikhail Tal, and Magnus Carlsen together in a bar, ideal for character group illustrations.
by Rakshit
Hidden Word Image Prompt: 'zsakib' in a Pride of Lions - AI Image Prompt
Hidden Word Image Prompt: 'zsakib' in a Pride of Lions
A creative prompt designed to test GPT Image 1.5's ability to embed a specific, hard-to-see hidden word ('zsakib') within a complex image—a pride of lions—that becomes obvious once spotted.
by Sakib
Aerial View of Landmark - AI Image Prompt
Aerial View of Landmark
A simple, direct prompt used to test GPT Image 1.5's ability to generate recognizable landmarks from an aerial perspective.
by PSS
2000s Film Camera Style Photo - AI Image Prompt
2000s Film Camera Style Photo
A simple style modifier prompt instructing GPT Image 1.5 to render the output image in the aesthetic of a 2000s film camera photo.
by a2/acc

準備好創作驚豔圖像了嗎?

立即使用這個提示詞開始生成令人讚嘆的AI圖像,或者瀏覽我們豐富的提示詞庫,獲取更多靈感。

立即生成 瀏覽所有提示詞