当前位置:AIGC资讯 > AIGC > 正文

AI绘画结合GPT 把Ai绘画与摄影玩明白

一、绘画与摄影有什么关系?

绘画和摄影是两种不同的艺术形式,它们都以其自身独特的方式捕捉和表达现实。在某些方面,它们是相互联系的,而在其他方面,它们又有所不同。​

相似之处:绘画和摄影都是创造性的表现形式,都需要艺术家具有审美观念和创新思维。它们都可以用来传达情感、观念,或者描绘人物、风景、生活场景等。它们都需要对色彩、构图、光线和阴影的敏感性和理解。​
不同之处:尽管两者都可以表现现实,但它们的创建方式有很大的不同。绘画是一种更直接的创作过程,艺术家可以完全按照自己的想象来创作,而不受任何限制。而摄影则需要捕捉现实中的瞬间,尽管后期处理可以改变一些元素,但基本的形象是由被摄物决定的。此外,绘画通常需要更长的时间来完成,而摄影可以迅速捕捉瞬间。
相互影响:在历史上,绘画和摄影的关系非常紧密。摄影的出现在19世纪中叶,挑战了绘画作为记录现实的主要方式,推动了艺术家寻找新的创作手法,这就导致了现代艺术的出现。另一方面,绘画也影响了摄影的发展,例如在构图、光线和色彩等方面。许多摄影师的作品都受到了绘画的影响。


总的来说,绘画和摄影是两种独特的艺术形式,它们既相互竞争,又相互影响。虽然它们在形式和技术上有所不同,但都是创造和表达美的方式

学习摄影词汇——摄影相关词汇整理

二、了解摄影师

这里用到了一个mj图书馆网站,里面有大量词汇

(可能需mofa)https://www.midlibrary.io/styles

从分类里面可以找到摄影师,可以从里面查看不同摄影师对应的风格,总共有300+

三、mj 摄影句式

通过GPT了解了摄影词汇和一些摄影大师,接下来就是把词汇组合 丢给midjouney 尝试不同的效果​

Midjourney 句式:​

<机位/角度>,<主体>,<环境>,<灯光>,<摄影风格/摄影作家> <其他> ​

先来个人物拍摄:​

特写,快照美学,超高清画面,一个美丽的女孩,18岁,在海边奔跑,蓝天白云,脸上带着微笑,超细节,宾得k1000,体积光,柯达彩色胶片拍摄,摄影师川内凛子 (Rinko kawauchi) ​

close-up,snapshot aesthetic,uhd image, A beautiful girl, 18 years old, running by the seaside, with blue sky and white clouds, a smile on her face, super detail,Pentax k1000,volumetric lighting,shot on kodachrome,art by Rinko kawauchi

动物拍摄:​

宠物摄影,360全景,樱花树下,落樱,茶杯贵宾犬,尼康单反相机,逆光,高清​

Pet photography,360 panorama,under the cherry tree,falling cherry blossoms,Teacup Poodle,Nikon SLR camera,backlight,high definition

Pet photography,35mm focal length,under the cherry tree,falling cherry blossoms,British Shorthair,Nikon SLR camera,backlight,high definition

四、让GPT发挥想象力帮你生成prompt​

这是参考网上midjourney的prompt 基础上 进行摄影模式的修改  核心部分在句式结构上 大家可以根据自己喜欢的句式进行调整​

Prompt 如下​

As a prompt generator for a generative AI called "Midjourney", you will create image prompts for the AI to visualize. I will give you a concept, and you will provide a detailed prompt for Midjourney AI to generate an image.​
Please adhere to the structure and formatting below, and follow these guidelines:​

- Do not use the words "description" or ":" in any form.​

- Do not place a comma between [ar] and [v].​

- Write each prompt in one line without using return.

Structure:​

[1] = [PROMPT]​

[2] = a detailed description of [1] with specific imagery details.​

[3] = a detailed description of the scene's environment.​

[4] = a detailed description of the scene's mood, feelings, and atmosphere.​

[5] = A photography style (e.g. highspeed photography, Time-lapse photography, DOF, Astrography, extreme long shot, Ultra-wide angle, Half-body shot, etc.) for [1].​

[6] = A description of how [5] will be executed (e.g. camera model and settings, DSLR,CanonEOSR5,SonyAlpha1.etc.)​

[7] = photography lighting(e.g. coruscating, Dappled, Glistening, Soft lighting, short lighting, Rim lighting, cinematic lighting, etc.)​

[8] = photographer(e.g. Mario Testino, Steven Meisel, Glistening,Patrick Demarchelier , Rinko kawauchi, etc.)​

[ar] = Use "--ar 16:9" for horizontal images, "--ar 9:16" for vertical images, or "--ar 1:1" for square images.​

[v] = Use "--niji" for Japanese art style, or "--v 5" for other styles.​

Formatting: ​

Follow this prompt structure: "/imagine prompt: [1], [2], [3], [4], [5], [6],[7],[8] [ar] [v]".​

Your task: Create 4 distinct prompts for each concept [1], varying in description, environment, atmosphere, and realization.​

- Write your prompts in [一只可爱的小狗].​

- Do not describe unreal concepts as "real" or "photographic".​

- Include one realistic photographic style prompt with lens type and size.​

- Separate different prompts with two new lines.​

Example Prompts:​

Prompt 1:​

/imagine prompt:A stunning Halo Reach landscape with a Spartan on a hilltop, lush green forests surround them, clear sky, distant city view, focusing on the Spartan's majestic pose, intricate armor, and weapons, Artwork, oil painting on canvas, --ar 16:9 --v 5

GPT生成的prompt:

然后输入了一只可爱小狗 再用GPT生成的prompt ​

A fluffy Samoyed puppy frolicking in a snowy landscape, the puppy's white fur contrasting with the stark white snow, pine trees dusted with snow in the background, a quiet, serene winter day, hyperrealistic style with attention to individual hairs and snowflakes, shot with a Canon EOS R5, using a telephoto lens to capture the dog from a distance, backlit by soft winter sunlight, reminiscent of the work by Rinko Kawauchi, --ar 1:1 --v 5

更新时间 2024-02-09