The author of this article is a design engineer of 360 Qi Dance Company
What is Midjourney?
Midjourney is an artificial intelligence program that can generate images based on text or pictures, and is currently set up on a Discord channel.
What is Discord?
Discord is a multi-functional social platform, the gameplay includes text, voice, video chat, in-game integration, etc.
To put it simply, we can think of Discord as WeChat, and Midjourney is an ai-generated applet in him.
How to use Midjourney: By sending the corresponding text (that is, the mantra we usually say) to the chat robot (midjourney bot) in the Discord channel, the chat robot returns the corresponding picture.
It should be noted that to use midjourney, you first need to surf the Internet scientifically, and then you need to register a discord account. Midjourney Discord : https://discord.gg/midjourney
There are many other tutorials on the Internet for the registration process of Midjourney, so I won’t repeat them in this article.
When we log in successfully, the following interface is displayed: We can create multiple servers according to our own needs, and add different applications to each service.
Below we create a new server and add the midjourney application to the server.
Click the "Add Server" button
Create it yourself (or use a template, the template will have some channels, set the template)
Just for me and my friends / for club or community use whichever you choose
Give my service a name and upload a cool avatar.png
This completes the creation of the server, click " Add your first APP " at the bottom
Come to the Discord App directory: enter Midjourney and press Enter
Select the first Midjourney Bot , open a new page, and click "Add to Server"
to continue
Now we have officially completed the operation of adding midjourney application in discord, and now we can start our journey of AI generating map!
Create your first AI artwork with Midjourney
Go back to our server, and then enter "/" in the input box below, it will pop up a series of commands, we choose to click "/imagine" to enter a set of keywords:
a cute baby of tiger
Press Enter, and after waiting for a while, Midjourney bot generates the following pictures. By
default, four pictures are generated. From left to right, from top to bottom, they are picture 1, picture 2, picture 3, and picture 4. The ratio of the pictures is 1:1.
U means to enlarge the picture, U1/U2/U3/U4 means to enlarge one of the four pictures respectively; V means to use the composition form of the picture to regenerate a group of similar pictures, V1/V2/V3 The order of /V4 is the same as that of U;
If you are still not satisfied with the current four pictures, then click the refresh button to generate the picture again.
We can also add some text (spells) to make the robot generate pictures more in line with your needs for pictures. It is best to follow the spell format below. There is no problem if one or two items are missing
The format of the mantra: subject + environment + composition lens + reference direction + image setting + parameters
a cute baby of tiger wearing a future space suit,the background is the universe, Natural light, realistic lighting and shading, pastel color, Whole-body perspective, tide play blind box,Pixar, the best picture quality and ultra-detail, 3D, C4D, Octane render, Ray Tracing,8K --ar 3:4 --niji 5
Let's break down the spell above
Main body : a cute baby of tiger wearing a future space suit: a cute baby of tiger wearing a future space suit
Environment : the background is the universe, Natural light, realistic lighting and shading, pastel color: the background is space, natural light, realistic lighting and shading, pastel color
Composition lens : Whole-body perspective,: Whole-body perspective
Reference direction : tide play blind box, Pixar,: tide play blind box, Pixar style
Image settings : the best picture quality and ultra-detail, 3D, C4D, Octane render, Ray Tracing, 8K: the best picture quality and ultra-detail, 3d style, c4d and oc rendering, ray tracing, 8k quality
Parameters : --ar 3:4 --niji 5 : Image ratio: 3:4, output mode niji5
Remember: a group of keywords needs to be followed by an English comma "," and then press the space bar, just leave a blank space, and there is a space between the parameters, no comma is required.
Let's take a look at what kind of pictures the robot will give us after entering the spell above? Isn't it amazing and interesting?
Some commonly used Imagine keyword suffix parameters
Aspect ratios (aspect ratios) --aspect, or --ar to adjust the ratio of the picture;
chaos (chaos) --chaos changes the multiplicity of results. Higher values produce more unusual and unexpected results;
Negative prompt (no) --no Negative prompt, adding --no at the end of the prompt word can make certain content not appear in the screen;
Version (version) Use the --version or --v parameters or use the /settings command and select the model version to use other models. Different models are good at processing different types of images.
--niji Another mod focused on anime-style images;
Next, I will share five sets of spells that are tried and tested, and everyone will start to use them for themselves!
01 - Gradient Overlay
![c43c6675bd38b820a3b7b0d7a21a7041.png](https://img-blog.csdnimg.cn/img_convert/c43c6675bd38b820a3b7b0d7a21a7041.png)
abstract image of waves created by computer, in the style of orange and blue, layered gestures, colorful installations, nature-inspired forms, uhd image, free-flowing lines, redshift --ar 16:9 --v 5.1
02 - space nebula wave
![a2372555c2f5f680490373e6e46accf9.png](https://img-blog.csdnimg.cn/img_convert/a2372555c2f5f680490373e6e46accf9.png)
Bokeh render, isometric view, unstable gravity space nebula waves, a timeless being made of luminescent viscous gel traveling faster than the speed of light through the event horizon of a blackhole --ar 16:9 --v 5.2
03-Snow Mountain Scenery
![95c15c573a978c984908300620888a15.png](https://img-blog.csdnimg.cn/img_convert/95c15c573a978c984908300620888a15.png)
wallpaper, a fantasy mountain, covered in snow, sunlight, waning light, bluesky, reinassance painting, highly detailed, intrincate, 8k --ar 16:9 --v 5
04-Simple and advanced gradient
![aaa4dd221360d82fd340dc83faff6654.png](https://img-blog.csdnimg.cn/img_convert/aaa4dd221360d82fd340dc83faff6654.png)
abstract background with blue streamers on light blue, simple, long exposure, blue, stripes, clear, Unreal Engine, industrial design, studio --ar 16:9 --v 5.2
05-Mech Warrior
![0839efae0809ec8052f007f5046a1839.png](https://img-blog.csdnimg.cn/img_convert/0839efae0809ec8052f007f5046a1839.png)
Big close-up mecha head, matte titanium exoskeleton, star war, space, futuristic, matte, right frontal shot, zoomed in center, ultra detailed, 8k, ultra realistic, cinematic atmosphere, unreal engine, octane render,--s 700 --v 4 --ar 16:9
In addition to the "/imagine" command, there are other commands that will be used. Let's briefly list them:
/ask Get answers to questions. You can ask some questions for midjourney to answer for you, similar to FAQ;
/blend blending images, a total of 6 images can be uploaded, and sending them to the robot will help you blend the uploaded images to generate a new set of images;
/docs is used in the official Midjourney Discord server to quickly generate links to topics covered in this user guide;
/fast Switch to fast mode, generally there is no need to switch this command for the duration of Fast use;
/help Displays useful basic information and tips about Midjourney Bot, help center, literally;
/imagine Use prompts to generate an image, this is the command to generate a picture, enter keywords to send;
/info View information about your account and any queued or running jobs, you can view related information such as the remaining drawing time of the account;
/stealth For Pro plan users ($60/month): switch to stealth mode. It means that the pictures you generate are not displayed in the community;
/piblic for pro plan users ($60/month): switch to public mode, literally;
10./subscribe generates a personal link for the user's account page;
11./settings View and adjust Midjourney Bot settings;
12./prefer option set Create or manage a custom option;
13./prefer option list View your current custom options;
14./prefer suffix specifies a suffix to add to the end of each prompt;
15./show regenerates the job within Discord using the image job ID;
16./relax Switch to relax mode. This mode is slower than Fast, and generally paid students will automatically switch to Relax after using Fast;
17./prefer remix Turn on/off the remix mode.
If you feel that keywords are difficult and you don’t know how to write them, don’t worry, I will recommend three useful keyword website AI prompt word (Prompt) tool libraries: https://prompthero.com/
Midjourney Keyword Assistant: https://prompt.noonshot.com/
ops prompt word studio https://moonvy.com/apps/ops/
Written at the end, the space is limited, I hope it will be helpful to those who have not used midjourne.
- END -
About Qi Wu Troupe
Qi Wu Troupe is the largest front-end team of 360 Group, and participates in the work of W3C and ECMA members (TC39) on behalf of the group. Qi Wu Troupe attaches great importance to talent training. There are engineers, lecturers, translators, business interface people, team leaders and other development directions for employees to choose from, and supplemented by providing corresponding training on technical skills, professional skills, general skills, leadership skills, etc. course. Qi Dance Troupe welcomes all kinds of outstanding talents to pay attention to and join Qi Dance Troupe with an open and talent-seeking attitude.