Midjourney V5 How To Upload A Reference Image Or Art And Use As A Prompt (2023)
Summary
TLDR该视频教程介绍了如何在Mid Journey V5中上传参考图片,并将其用作提示以生成AI图片。主要步骤是:在Discord上上传参考图片,复制图片链接;在Mid Journey的提示框中输入所需图片的描述,并粘贴参考图片链接;在链接后添加--IW 2.0,以最大化使用参考图片;等待Mid Journey生成结果图片。该教程简明扼要地概述了如何利用Mid Journey V5的图像引导功能生成高质量AI图片。
Takeaways
- 📥 首先需要将要作为提示的参考图片发送到Discord并复制图片链接
- 👉 在Mid Journey发送/Imagine命令并在提示框输入你想要的图像描述
- 📎 在描述后粘贴参考图片的链接
- ➖ 输入两个破折号再输入IW和一个数值来设置图片权重
- 🔢 IW数值范围0.5-2.0,数值越高生成的图像越贴近参考图片
- ☑️ 输入提示并等待Mid Journey生成图像
- 🎨 Mid Journey会生成多个版本,选择效果最好的图像
- 👍🏻 在网页上放大选中的图像并下载保存
- 🖼️ 保存的图像可以用于任何地方
- ✅ 这就是如何在Mid Journey中使用参考图片作为提示的方法
Outlines
🐶 如何在Mid Journey中上传参考图片
本段介绍了在Mid Journey中如何上传参考图片或艺术品,并将其用作提示的步骤。需要先在Discord中上传要参考的图片,右键复制图片链接。然后在Mid Journey的提示框中,描述所需的图像,并粘贴参考图片的链接。最后设置image weight参数为2.0,以尽可能贴近参考图片生成结果。
📸 生成的图片进行调整和保存
本段展示了Mid Journey根据提示所生成的4张图片效果。选择了效果最好的2张图片进行调整,然后保存到电脑上供进一步使用。总结了如何在Mid Journey中使用参考图片生成所需图像的全过程。
Mindmap
Keywords
💡midJourney
💡生成图像
💡上传参考图片
💡提示
💡图像权重
💡生成参数
💡风格化
💡质量
💡背景
💡实用教程
Highlights
在Discord中打开mid Journey机器人
右键点击上传的图片,选择复制链接
在提示框中输入想要生成的图像描述
粘贴参考图片的链接
添加双破折号和IW表示图像权重
图像权重决定参考图片的重要程度
图像权重范围0.5到2.0,数值越高关注度越高
设置2.0使生成图片尽可能贴近参考图片
mid Journey生成4张类似狗的图片
选择效果最好的图片进行后续处理
下载并保存喜欢的图片
上传参考图片可指导mid Journey生成所需图片
调整图像权重参数控制生成图片的相似程度
这个教程详细解释了mid Journey的参考图片使用
欢迎提出任何问题和建议
Transcripts
mid Journey version 5 how to upload a
reference image or art and use it as a
prompt so go ahead and open Discord to
access the mid Journey bot and the first
thing that we need to do is to send here
the images that we want to be referenced
so open the folder where it is located
make sure that it's downloaded on your
device so here is a dog which I got from
drag it to Discord and present enter to
send it once it's uploaded right click
on it and among these additional options
select copy link
now that we have its link let's Now
command mid Journey so type in slash
followed by Imagine and click on this
on the prompt box just type in what you
want to make in your image so for
example I want the same dog to look up
so I'll just type in dog looking with
heads up and then I'll also add natural
light
trees as its background and then I'll
just add a photorealistic so that it
will be high quality
now press command V or Ctrl V on your
keyboard to paste the link of the photo
that we copied a while ago
after that type in Double Dash so dash
dash followed by the letters IW which
means Image Weight image weight on Mid
journey is the parameter that sets the
weight of the image prompt relative to
the text while generating AI images its
value can range from 0.5 to 2 with 0.5
being the lowest and 2.0 is the highest
now I want it to be copied as close as
possible to the reference image so I'll
add in space and type 2.0 enter that and
then wait for my journey to finish
generating your image now as you can see
it generated four images of the same dog
but its head and eyes are looking up
because of what we typed in here upscale
what you want as for me the 1 and 2 is
the best result so let's use that open
in the web and that's it you can
download and save it to your computer
and use the image to anywhere you want
so that is how to upload a reference
image or art and use it as a prompt in
mid Journey version 5. do let us know if
this video helped you out and if you
have any question feel free to drop them
down below thank you for watching and
have a nice day
浏览更多相关视频
![](https://i.ytimg.com/vi/uvUOcHDoBkU/hq720.jpg)
AI Face Studioを使って、DeepFake画像を生成する方法!ディープフェイク以外にも、ステッカーやドット絵に変換できる素敵なサイト😆
![](https://i.ytimg.com/vi/yZQoJg2RCKI/hq720.jpg)
Block reference parts of images just like text with Obsidian Excalidraw
![](https://i.ytimg.com/vi/tndR-onvxY0/hq720.jpg)
Stylar AI Tutorial - New AI for Text to Image & Image to Image
![](https://i.ytimg.com/vi/7YSLejDsmh8/hqdefault.jpg?sqp=-oaymwEXCJADEOABSFryq4qpAwkIARUAAIhCGAE=&rs=AOn4CLCZOlKZaqYc3zuKtLbRswtQdYUlUQ)
【AI賺錢喂飯級教程】小紅書用AI生成超萌爆火柴犬視頻,一周輕鬆漲粉過萬的秘訣大公開,新手也能輕鬆月入5萬+!
![](https://i.ytimg.com/vi/uHFd0XoHRxE/hq720.jpg)
Excalidraw-Obsidian 2.0.15: Image Cropping, Masking and Metadata
![](https://i.ytimg.com/vi/MTdbhePtCco/hq720.jpg)
Mastering Concept Visualizations: A Simple Workflow for Creating Effective Visuals
5.0 / 5 (0 votes)