AI繪圖筆記 使用ChatGPT 協助產生Stable Diffusion 正向提示詞Positive Prompt

Stable Diffusion 提示詞是指用來描述所需生成圖像的文字，例如 “1girl”、”1boy” 等，AI 會根據這些描述生成相應的圖片。Stable Diffusion 提示詞又稱為關鍵詞、提示、咒語；簡單說就是用來跟Stable Diffusion溝通的方式。

但如何將中文描述的風格轉為提示詞？可以讓火紅的ChatGPT來幫助。

例如輸入以下要求：

我在使用一個名為 Stable Diffusion 的 AI 繪圖工具。指定你成為 Stable Diffusion 的提示生成器。接著我會在想要生成的主題前添加斜線（/）。你將在不同情況下用英文生成合適的提示。例如，如果我輸入 /一個亞洲女孩在草地上的美麗形象 ,您將生成prompt用以下的格式：

1girl, solo, asia, sitting, an extremely delicate and beautiful, beautiful detailed sky, beautiful detailed eye, side blunt bangs, hairs between eyes, close to viewer, breeze,flying splashes,flying petals, wind, grass

The prompt requires the following elements. 

(1) Words to describe the method of depiction, such as digital painting, oil painting, realistic 8k.

(2) Words to describe external features, ornaments and belongings (also specify colors, patterns, shapes), such as button-up blouse, hair tied back, red lipstick.

(3) Words to describe age, body type, hair color, hairstyle, hair length, hair accessory, eye color, eye shape, facial expression, breast size, and clothing. 

(4) Words to describe background details, such as inside room, starry sky, forest, riverside. 

(5) Words to direct the pose from head to toe can also be added to the prompt. Be sure to include "1girl, solo," at the beginning of the prompt. Your suggested prompt should be plain text and no use of 「」or "" or line breaks. 

Please describe each elements of the prompt after you have suggested your prompt. Prompts may use danbooru tags. Based on these, please tell me a 150-element prompt that can generate illustrations everytime i asked.

接著只要輸入

/在操場上認真練習的運動選手

就會得到：

1athlete, solo, running, focused expression, sportswear, sneakers, ponytail, sweating, outdoor

This prompt includes: 

(1) Depiction method: None specified. 

(2) External features: The athlete is wearing sportswear and sneakers. She has a ponytail and is sweating. 

(3) Age, hair, facial expression, clothing: None specified, except for the ponytail and sweating mentioned in the external features. 

(4) Background details: The athlete is outdoors, specifically on a running track or in a sports field.
(5) Pose: The athlete is running and has a focused expression.

這樣對於基礎的描繪就可以完成。

通常需要有高畫質產出，可以加上這些：

(8k, RAW photo, photorealistic:1.25), masterpiece, best quality, highres, looking at viewer, highly detailed background, (photo realistic:1.2), (MASTERPIECE), (ULTRA DETAILED), (PHOTO-REALISTIC), SHARP, HDR

正向提示詞Positive Prompt範例：

/在圖書館看書的女孩

1girl, solo, digital painting, sitting, glasses, book, library, long hair, brown hair, loose hair, hair between eyes, peaceful expression, casual clothing, indoor, shelves, table, chair, natural light, calm atmosphere, bookish, (8k, RAW photo, photorealistic:1.25), masterpiece, best quality, highres, looking at viewer, highly detailed background, (photo realistic:1.2), (MASTERPIECE), (ULTRA DETAILED), (PHOTO-REALISTIC), SHARP, HDR

 有任何問題歡迎留言討論 !!

更多Stable Diffustion AI繪圖 討論文章：

• 環境安裝 藉由Google Colab 來安裝Stable Diffusion 玩AI繪圖
• 基本功能txt2img AI繪圖筆記 使用ChatGPT 協助產生Stable Diffusion 正向提示詞Positive Prompt
• 基本功能txt2img Stable Diffusion AI繪圖 萬用負面關鍵字Negative Prompt
• 如何使用LoRA AI繪圖筆記 安裝完基本的模型後 如何藉由套用客製化LoRA 幫上原亞衣穿上婚紗