Midjourney Prompts, Formulas & Tips (for STUNNING Ai Art)

Wade McMaster - Creator Impact
21 Sept 202306:29

TLDRThe video script discusses various prompt formulas for creating unique images using AI, such as blending emotions or objects with styles, contrasting styles, and object placements. It introduces style stacking, where multiple styles are layered, and chain prompting, which involves building on a sequence of images. The script also covers detailed prompts that specify subjects, backgrounds, styles, and parameters for more controlled outcomes. The video encourages viewers to experiment with these methods for creative image generation.

Takeaways

  • πŸ“ Start with simple prompts like emotions or objects to create interesting images using AI's interpretation.
  • 🎨 Use the 'object and style' formula to combine a simple object with a strong style for unique results.
  • 🌟 Explore 'contrasted styles' by merging two conflicting styles or 'contrasted objects' by placing an object in an unexpected setting.
  • πŸ”₯ Experiment with 'style stacking' to layer multiple styles and elements on top of a base prompt for increasingly complex images.
  • πŸ”— Utilize 'chain prompting' by taking an image you like and using it as a reference for further image prompts.
  • πŸ‘Ύ Be open to AI's creative interpretations, even if the final image isn't exactly what you envisioned.
  • 🎭 For more control, use a detailed prompt including main subject, details, background, art style, lighting, colors, and parameters.
  • πŸ“Š Adjust parameters like aspect ratio and stylize settings to fine-tune the output.
  • πŸ’‘ Consider signing up for newsletters or downloading cheat sheets for guidance on using AI art tools effectively.
  • πŸ”„ Remember that you can't always re-roll and change your prompt; use 'remix mode' to add more elements or tweaks.
  • 🌐 Share your AI-generated images with others to inspire their own creative prompts and explorations.

Q & A

  • What is a blunt prompt in the context of the video?

    -A blunt prompt refers to a simple one or two-word input, such as an emotion or an object, that is used to create an image based on the AI's interpretation.

  • How does style stacking work in image creation?

    -Style stacking involves taking a simple prompt and adding multiple styles or elements to it, building upon the initial image to create a more complex and detailed result.

  • What is chain prompting and how is it executed?

    -Chain prompting is a technique where an image is created, then upscaled, and used as a reference in subsequent image prompts, allowing for a continuous evolution of the image based on the AI's interpretation of the previous result.

  • What are contrasted styles and how do they enhance image creation?

    -Contrasted styles involve combining two conflicting styles or themes, such as traditional and modern, to create a unique and interesting image that blends the contrasting elements together.

  • How can one utilize a very detailed prompt in image creation?

    -A very detailed prompt involves specifying the main subject, details of subjects, background, art image style, lighting, colors, and parameters to take more control over the image creation process, resulting in a more predictable and desired outcome.

  • What is the purpose of using a parameter in image creation?

    -Parameters are used to fine-tune the specific qualities of the generated image, such as aspect ratio, stylization, and other settings that can enhance or alter the final result.

  • How does the remix mode work in image prompting?

    -The remix mode allows users to add additional elements or adjustments to the image by typing '/settings remix' which enables further customization and creativity in the image generation process.

  • What is the significance of the DMT style mentioned in the video?

    -The DMT style refers to a distinctive visual aesthetic often associated with the psychedelic and surreal imagery that can be inspired by the effects of the DMT substance. It is used as an example of a strong style that can be applied to objects or subjects in image creation.

  • How does the video suggest using MidJourney for image creation?

    -The video suggests using MidJourney as a tool to create images by inputting various prompts and parameters, allowing users to explore their creativity and generate unique images by blending styles, subjects, and settings.

  • What is the role of the AI in interpreting prompts for image creation?

    -The AI plays a crucial role in interpreting the prompts by analyzing the input and generating images based on its understanding of the given instructions, styles, and themes, adding a layer of unpredictability and creativity to the final result.

  • What advice does the video give on managing expectations with AI-generated images?

    -The video advises that while detailed prompts can help achieve more predictable results, there is always an element of unpredictability in AI-generated images. It encourages users to embrace the creativity and sometimes unexpected outcomes of the AI's interpretation.

Outlines

00:00

🎨 Exploring Blunt Prompts and Object-Style Combinations

This paragraph discusses the concept of blunt prompts, which are simple one or two-word prompts that can evoke an emotion or represent an object. The speaker explains how these prompts can lead to interesting images based on the AI's interpretation. The paragraph then transitions into discussing more complex prompt formulas, starting with simple object-style combinations, where a basic object, like a mythical creature, is paired with a distinctive style, such as 'DMT style.' The speaker also mentions the potential for contrasting styles and objects, combining traditional and modern elements to create unique images.

05:01

πŸ”₯ Advancing with Style Stacking and Chain Prompting Techniques

The speaker introduces 'style stacking,' a method where a basic prompt is layered with additional styles, resulting in a more intricate and detailed image. An example given is 'Pikachu as the Terminator,' which can then be further enhanced with styles like 'DMT style' or '3D render.' The paragraph continues with 'chain prompting,' a process where an initial image is used to generate subsequent images by adding new elements and remixing the prompt. This technique allows for a continuous evolution of the image, and the speaker notes the creative potential in sharing these images for others to use as a starting point for their own chain prompts. The paragraph concludes with an acknowledgment and appreciation for Future Tech Pilot, who introduced the speaker to this method.

Mindmap

Keywords

πŸ’‘Blunt Prompt

A 'blunt prompt' refers to a straightforward and simple input given to an AI system to generate an image. In the context of the video, it is used to describe the initial and basic way of interacting with the AI, where the user inputs a single word or an emotion to see how the AI interprets and visualizes it. An example from the script is using a single word like 'mythical beast' to create an image.

πŸ’‘Object and Style

The 'object and style' concept is a technique where an object, often a mythical or fantastical one, is combined with a distinctive artistic style to produce an image. This method is showcased in the video as a way to create more intricate and visually interesting outputs by adding a style descriptor, such as 'DMT style', to a simple object, resulting in a unique visual blend.

πŸ’‘Contrasted Styles

Contrasted Styles involve pairing two conflicting or contrasting styles to create an image with a unique aesthetic. The video describes using traditional art styles like 'Yukio-a' and modern techniques like '3D render' to juxtapose old and new, resulting in a fresh interpretation of a subject.

πŸ’‘Contrasted Objects

In the context of the video, 'Contrasted Objects' refers to the creative process of placing an object in an environment that is not typically associated with it. For instance, placing a 'fish in the desert' creates a surreal and unexpected image, challenging the viewer's expectations and prompting imaginative interpretations.

πŸ’‘Style Stacking

Style stacking is a method where multiple styles are layered on top of a base image or concept to create a complex and rich visual. The video provides an example of this technique by suggesting 'Pikachu as the Terminator DMT style 3D render', which builds upon the initial concept by adding various stylistic elements and transformations.

πŸ’‘Chain Prompting

Chain prompting is a process where an image generated by the AI is used as a starting point for further image generation. The user takes a generated image, finds an aspect they like, and then uses it as a reference to create a new image, continuing this cycle to explore different creative paths. This technique is highlighted in the video as a way to iteratively refine and explore the AI's capabilities.

πŸ’‘Detailed Prompt

A 'detailed prompt' is an advanced method of interacting with the AI that involves providing comprehensive instructions to control the output more precisely. This includes specifying the main subject, details, background, art style, lighting, colors, and other parameters. The video demonstrates this by constructing a prompt that results in an image of a 'samurai warrior in red armor, battle damaged, on the beach during sunset, in 3D Unreal Engine style with cinematic lighting and warm tones'.

πŸ’‘Mid Journey

In the context of the video, 'Mid Journey' appears to be the name of the AI system being used for image generation. It is described as an AI that can interpret user inputs, known as 'prompts', and produce images based on those inputs. The system allows for a range of creative exploration, from simple blunt prompts to more complex style stacking and detailed prompts.

πŸ’‘Cinematic Lighting

Cinematic lighting refers to the use of lighting techniques in image creation that mimic those used in filmmaking to evoke specific moods or atmospheres. In the video, it is mentioned as one of the parameters that can be adjusted in a detailed prompt to achieve a desired visual effect, such as 'golden hour' lighting to create a warm, sunset ambiance.

πŸ’‘Unpredictability

Unpredictability in the context of the video relates to the inherent variability in AI-generated images. Even with detailed prompts, the AI's interpretation can lead to unexpected results, which can be a source of surprise and creativity. The video encourages embracing this aspect, comparing it to hiring an artist who sometimes delivers results that are not exactly as envisioned but are still interesting and valuable.

πŸ’‘Remix Mode

Remix mode is a feature within the AI system that allows users to modify and add elements to existing prompts or images. The video describes it as a way to continue the creative process by making adjustments and exploring new combinations, such as adding 'massive fire' to an image to see how the AI interprets and visualizes the new elements.

Highlights

The use of blunt prompts to create interesting images by combining a simple object or emotion with an AI's interpretation.

The concept of 'simple object and style' prompt formula, where a mythical beast is combined with a strong style like DMT style to produce unique images.

The introduction of 'contrasted styles' where two conflicting styles are merged, such as traditional Yukio-A with a modern 3D render twist.

The 'contrasted some object' method, which involves placing a completely different object in an unexpected setting or combining it with a contrasting style.

The technique of 'style stacking', where a basic prompt like 'Pikachu as the Terminator' is layered with additional styles or elements like 'DMT style 3D render'.

The ability to iteratively refine and build upon an image by stepping back and adding to it in style stacking, allowing for a more personalized outcome.

The explanation of 'chain prompting', a method where an initial image is upscaled and used as a reference for further image prompts, creating a chain of evolving images.

The process of upscaling and copying the link address of an image to use in subsequent image prompts for continuous creative development.

The concept of 'very detailed prompt' which involves specifying the main subject, details of subjects, background, art image style, lighting, colors, and parameters for more control over the AI-generated image.

The recommendation to use Mid Journey for its ability to produce creative and sometimes unexpected results, akin to hiring your own artist.

The advice to not always expect the exact image envisioned and to embrace the AI's unique interpretation for a more creative experience.

The mention of a parameter cheat sheet available for download and the option to receive it by signing up for an email newsletter.

The reminder that remix mode can be turned on by typing '/settings' followed by 'turn on remix' to add more elements and continue tweaking the image prompts.

The encouragement to combine different prompt formulas for better results and to explore the creative potential of AI image generation.

The acknowledgment and credit given to Future Tech Pilot for introducing the chain prompting method.

The closing statement encouraging viewers to like the video if they found it useful and the anticipation of future content.