Master Midjourney Prompts: A Guide to creating creative Midjourney Prompts with Chat GPT Assistance
Boost your imagination with ChatGPT and Midjourney
The ability to illustrate and create images is something that I have always wanted, but unfortunately, it is not a native skill for me. Although I studied it at university and have some basic skills, I need to be more skilled to illustrate in high quality.
Since AI image generation tools have been released, this is the right time to start creating illustrations. Since I no longer need painting skills, I only need imagination and prompt engineering skills to write text prompts that create the perfect image.
In contrast to writing prompts, which are more technical tasks, imagination is a much more creative skill we need to develop to succeed. Imagination can take time to develop and doesn’t always come to mind immediately.
Imagination is like cooking on low heat; bringing out the best flavors takes time and patience. Sometimes, you must keep heating the ingredients until they are just right; it is the same with imagination. You must keep adding and refining ideas until you get the perfect idea or image.
In this post, I want to share with you a technique I use to enhance my imagination and create better images. The technique involves using ChatGPT to open up my imagination.
ChatGPT helps me think outside the box and come up with creative ideas and enables me to explore new ideas and express my thoughts and feelings more accurately.
The article also included gifs to help you understand the process better.
Why should we use ChatGPT for the Imagination process?
Speed and Efficiency: ChatGPT generates ideas faster than human thinking. With this speed, we can come up with multiple creative ideas quickly, giving us more options to choose from.
A serendipitous discovery: One of the best things about using ChatGPT for imagination is that it helps us develop ideas we wouldn’t think of. With random but relevant suggestions, artificial intelligence helps us develop new creative ideas.
For instance, imagine that you want to create an image that shows a group of people playing volleyball on the beach. You can ask ChatGPT to add a random idea to the text. For example, this text prompt:
Please provide more details in this text. The idea should be crazy. Group of people playing volleyball on the beach.This is the result:
A group of people play beach volleyball on a sun-kissed beach with a giant pink flamingo that stand on the sand hit the ball near the net.
Let’s be honest. How easily do you think about the giant pink flamingo that joined the game?
Working with the machine is the key
Many people assume that using AI tools is easy because the tool does all the work for them. They think it is enough to write a simple prompt and they can get the desired results.
From my point of view, it is not true. It is essential to actively collaborate with the machine and experiment with it to achieve desired results.
Remember that you make the decisions, not the machine; the AI is here to help you think faster and generate ideas more efficiently, but you make the final decisions.
The idea is to play “ping-pong” with the machine, engaging in a back-and-forth dialogue to refine your ideas and arrive at the desired outcome. You can adjust the results yourself or ask the machine for assistance, but it is crucial to understand what you want.
Let’s take the example of the people that play volleyball that I explained above. The result was:
A group of people play beach volleyball on a sun-kissed beach with a giant pink flamingo that stand on the sand hit the ball near the net.Apart from that, You can add information manually, like the ball size (big, small) or the ball color (green, white). With the iterations, you can find the most accurate results for your needs.
Here is a variation of the midjourney prompt that I tweaked:
A group of people play beach volleyball on a sun-kissed beach with a giant green turtle that stand on the sand and hit the ball near the net.
Where did I get the idea for this technique?
I enjoy listening to podcasts because they provide a way to learn and discover creative ideas while I do other tasks like walking, cleaning, or taking the metro. I am curious by nature, so finding new things excites me.
In one podcast, I listened to a talk with an illustrator that explained his creative process. Unfortunately, I cannot recall the specific podcast. However, the illustrator’s approach significantly impacted how I now work with AI tools like Midjourney and ChatGPT.
Here I explained the technique he described so you can understand the root of the technique with the AI tools.
The illustrator starts his creative process with a general idea and asks himself for more details about the image. For example, he considers the location, the people or objects in the image, the time of day, the colors, the foreground and background elements, and how all the elements within the image interact with each other.
Applying this method helps him imagine better and understand what he wants to create in the image. The process focused his imagination and helped him create the final image.
For instance, if he started with a concept like “a person in a city,” he would ask himself a series of questions to flesh out the scene:
- Which city?
- Where in the city?
- What is the person wearing?
- What time of day is it?
- Is it raining?
- What is the person doing?
- Are there people nearby?
This process of questioning and imagining helps refine ideas and guides the illustrator in creating the image.
By applying this same process to AI tools, you can develop better visuals and utilize the tools’ potential to increase your imagination.
In the creative process, mistakes are important
People often express what they don’t want instead of focusing on their desires. Exploring possibilities and sometimes making mistakes is necessary to understand what you want. More than that, mistakes can lead to surprising discoveries and innovative ideas.
For example, searching for a rental house can take time and effort. Even if you have a general idea of where you want to live, finding the right fit may take some exploration. You may need to see several places before finding the one that fits your needs.
When working on a concept, it is essential to remember that we may only sometimes have a clear idea of what we want.
Before arriving at the desired outcome, you should experiment with various sketches and ideas, and the mistakes you make can often be a valuable source of innovation and creativity.
The microwave oven is one example of innovation that resulted from a mistake. In 1946, an engineer named Percy Spencer was working on a magnetron when he noticed a candy bar in his pocket had melted. This led to his discovery that microwaves could be used to heat food, and the next year the first commercial microwave oven was released.
However, since computers are generally programmed to be accurate, mistakes can be challenging (Don’t worry, ChatGPT will be more accurate in the future). Because of that, it is crucial to encourage the AI to venture outside the box to achieve unexpected results.
You can push AI to produce more diverse and creative outputs by intentionally introducing errors or unconventional ideas into your prompts.
It’s all about embracing uncertainty and making mistakes, so you can unlock new possibilities and get better results that surprise you. For example, instead of giving AI a simple command like:
A forest with trees
You could add an element of surprise by saying:
A forest filled with singing trees that put on a musical performance every day
The technique
You can use the technique in two ways. One way is to provide ChatGPT with a prompt and ask it to offer ideas on completing it. From there, you can iterate by asking ChatGPT to add or remove information.
The second technique is to give ChatGPT a simple prompt with different parameters you want it to include and then iterate over it. Both techniques have the same idea but are different. Let’s look at each step by step.
Option 1: Provide ChatGPT with a prompt and ask it to offer ideas
By engaging in a back-and-forth conversation with ChatGPT, you can refine your ideas and obtain a more accurate image description to input into Midjourney.
To make it easier to understand, I added the steps you should take below. Do not take it as a recipe; play with it and tweak it to develop your prompting skills.
Start with a simple prompt
Begin with a simple description of the scene you want to create. For example:
A person drinks coffee.The image Midjourney generated
Add details using ChatGPT
To make your image more precise and help you discover more, use ChatGPT to provide details to your description. You can ask specific questions or request more information to create an effective prompt.
For instance, you can use this prompt and run it in ChatGPT:
Add some details to the text: "A person drinks coffee"
Here is the result:
In the cozy atmosphere of a bustling café, a young woman with auburn hair and glasses sits by the window, sipping her steaming cup of freshly brewed coffee. The delicate aroma of the dark roast fills her senses as she savors each sip, momentarily escaping the pressures of her daily life. With every passing moment, she feels the invigorating warmth of the coffee coursing through her veins, rejuvenating her energy for the tasks ahead.The image Midjourney generated
I asked ChatGPT to change the location
As you receive responses from ChatGPT, you can refine your prompt by asking for additional details, cutting unnecessary information, or making manual adjustments.
This iterative process helps you describe the scene you want to create.
For example, if you want, you can ask ChatGPT to add a place where the person drinks the coffee.
Here is the result:
Seated comfortably in her sunlit kitchen, a young woman with auburn hair and glasses takes a moment to herself, sipping her steaming cup of freshly brewed coffee. The delicate aroma of the dark roast fills her senses as she savors each sip, momentarily escaping the pressures of her daily life. With every passing moment, she feels the invigorating warmth of the coffee coursing through her veins, rejuvenating her energy for the tasks ahead.ChatGPT changed the location from “cozy atmosphere of a bustling café” to “Seated comfortably in her sunlit kitchen”. Be aware that the text is not 100% the same, so if you want to be 100% accurate, you can make changes manually.
This is the image I generated, I only changed the location manually.
Prompt
Seated comfortably in her sunlit kitchen, a young woman with auburn hair and glasses sits by the window, sipping her steaming cup of freshly brewed coffee. The delicate aroma of the dark roast fills her senses as she savors each sip, momentarily escaping the pressures of her daily life. With every passing moment, she feels the invigorating warmth of the coffee coursing through her veins, rejuvenating her energy for the tasks ahead.The image Midjourney generated
Experiment with unconventional ideas
Don’t be afraid to introduce surprising or unconventional elements into your prompt to unlock your creativity and generate unique images. You can develop innovative ideas that make your image stand out by embracing the unexpected.
For example, you can ask to chat to add something like:
Add to the text a crazy detail.Once satisfied with your image description, input it into Midjourney to create your image.
Option 2: Generate unique prompts with ChatGPT’s parameter list
ChatGPT can generate a detailed list of parameters so you can build your prompt template. Then, you can request ChatGPT to fill out the parameters simultaneously and get a different prompt every time you can run in Midjourney.
Ask for the parameters list
Run this prompt to get a parameters list.
Prompt example:
I want to describe an image of a person drinks coffee. Please give me a list of 15 parameters so I can better describe the image. For example, where and what?
ChatGPT generated this list:
1. Location: Describe where the person is drinking coffee (e.g., at a cafe, at home, in a park).
2. Time of day: Indicate if it's morning, afternoon, or evening.
3. Posture: Describe the person's body posture (e.g., sitting, standing, leaning).
4. Hand holding coffee: Specify which hand the person is using to hold the coffee (e.g., left or right).
5. Type of cup: Describe the cup or mug the person is using (e.g., ceramic, paper, glass, color).
6. Coffee type: Specify the type of coffee (e.g., espresso, latte, cappuccino, black coffee).
7. Facial expression: Describe the person's facial expression (e.g., happy, content, thoughtful, focused).
8. Clothing: Mention the clothes the person is wearing (e.g., casual, formal, specific colors or patterns).
9. Hairstyle: Describe the person's hairstyle (e.g., short, long, curly, straight, updo).
10. Surroundings: Mention any notable features in the background (e.g., plants, art, other people, furniture).
11. Lighting: Describe the lighting in the scene (e.g., natural sunlight, dim, artificial, warm).
12. Weather (if outdoors): Mention the weather conditions (e.g., sunny, cloudy, rainy, snowy).
13. Accessories: Note any accessories the person might be wearing (e.g., glasses, hat, scarf, jewelry).
14. Additional objects: Mention any other objects the person might be interacting with (e.g., newspaper, smartphone, book).
15. Age and gender: Describe the person's approximate age and gender (e.g., young adult female, middle-aged male).Create a template for ChatGPT
Write what you want ChatGPT to generate, add a list of parameters and an example of the output. Then ask ChatGPT to generate a final prompt.
It’s not necessary to select all the parameters, just the ones you want.
Here is an example:
Act like an illustrator and write an image description based on the PARAMETERS to describe an image in a detailed manner.
-Add your ideas for each parameter.
-The writing style should be like this PROMPT EXAMPLE.
-Write only the description.
PROMPT EXAMPLE: A young adult is sitting in a cozy coffee shop in a hip area, sipping a creamy cappuccino and working on their laptop. They're wearing casual clothing and have a focused expression on their face, typing away at their computer. The image represents mid-morning, with the sun shining brightly outside.
PARAMETERS=
·Location: Describe where the person is drinking coffee (e.g., at a cafe, at home, in a park).
·Time of day: Indicate if it's morning, afternoon, or evening.
·Posture: Describe the person's body posture (e.g., sitting, standing, leaning).
·Hand holding coffee: Specify which hand the person is using to hold the coffee (e.g., left or right).
·Type of cup: Describe the cup or mug the person is using (e.g., ceramic, paper, glass, color).Ask ChatGPT to write a prompt
Input the prompt into ChatGPT and let it generate an image description.
Here is the result from the prompt I showed above:
An individual is enjoying a warm cup of coffee in a serene park filled with blooming flowers and lush greenery. The scene captures the late afternoon ambiance, with the sun setting, casting a golden hue on the surroundings. The person is leaning against a park bench, one leg crossed over the other, exuding a relaxed vibe. They hold their coffee cup in their right hand, taking occasional sips as they admire the scenery. The coffee is contained in a reusable, stainless steel travel mug adorned with an intricate design, reflecting both style and environmental consciousness.
Input the detailed prompt into Midjourney
Input the detailed prompt into Midjourney to generate an image. Feel free to cut and tweak the result.
The image Midjourney generated
More tips
- If you want to tweak the generated prompt or make small changes, you can edit the description manually or ask ChatGPT to make the desired adjustments. Be aware that ChatGPT may change more than you anticipated when making adjustments, so manual editing might be necessary to achieve the desired result.
- When certain parameters must be included, specify them at the prompt before inputting them into ChatGPT. This way, the AI will maintain those specific parameters while generating the rest of the description.
- Remember that the key to generating more accurate and creative images using Midjourney and ChatGPT is actively collaborating with the AI, experimenting with different parameters, and refining your ideas until you achieve the desired outcome.
- You can also add Midjourney different styles parameters to the prompt, like illustration style, art style, aspect ratio, lighting style, and materials. This will give you additional control over your designs, allowing you to create more original visuals.
To summarize
In this article, I showed you how to develop your imagination using ChatGPT and Midjourney to become more creative.
I explained that I enjoy illustration, but I don’t have natural drawing abilities, and why I believe anyone can become an illustrator today with imagination and prompt engineering skills.
Next, I shared how I came across this technique. I listened to an illustrator explain his process for creating images and how he uses his imagination. Additionally, I talked about the importance of making mistakes and why we should allow the computer to make mistakes in the process.
Then, I explained the technique I use. I showed two ways: one where we give ChatGPT a simple prompt and ask it to develop it further by adding details, and another where we make ChatGPT generate a list of parameters to generate prompts.
More about AI tools
If you liked this article, you should check out my other articles about AI.
Thank you for reading the article. I hope it helped you understand how to use Midjourney to generate many prompts that will allow your imagination to explore what image you want to create. Please feel free to share it with your friends or team members, and if you have any questions, please let me know.
If you enjoyed my article, I suggest you follow me and subscribe so you’ll receive an email whenever I post.
