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

Wade McMaster - Creator Impact
21 Sept 202306:29

TLDRThe video script discusses various methods to enhance AI art creation using Midjourney, a powerful AI art generator. It starts with simple prompts, such as an emotion or object, and then delves into more complex formulas like combining an object with a specific style, contrasting styles, and object placement in unusual settings. The concept of 'style stacking' is introduced, where multiple styles are layered on top of a base prompt, allowing for a more intricate and personalized result. 'Chain prompting' is another technique where an image is upscaled and used as a reference for further prompts, creating a chain of increasingly detailed images. The video also mentions a detailed prompt formula that includes the main subject, background, art style, lighting, colors, and parameters to give the user more control over the final image. The speaker emphasizes the importance of allowing some creativity to the AI, comparing it to hiring an artist. The video concludes by encouraging viewers to experiment with different prompt formulas and to embrace the unique results that Midjourney produces.

Takeaways

  • 🖌️ Simple prompts can generate interesting images based on AI's interpretation, referred to as 'blunt prompts'.
  • 🎨 Combining an object with a specific style, like 'mythical beast in DMT style', can produce unique visuals.
  • 🤖 'Contrasted Styles' involve merging two conflicting styles or placing an object in an unexpected setting for a modern twist on traditional art.
  • 🔀 'Style Stacking' is a method where you layer multiple styles on a single prompt, allowing for complex and creative results.
  • 🔄 'Chain Prompting' involves using an existing image as a reference to create a new image, which can then be used to generate further images.
  • 🔍 For detailed control, include main subject, details, background, art style, lighting, colors, and parameters in your prompt.
  • 📐 Using parameters like aspect ratio and stylization can help refine the output to match your vision more closely.
  • 🧩 It's possible to remix and combine prompt formulas for more nuanced and specific results.
  • ⚖️ While you can guide the AI, there's an element of luck and the AI's interpretation, so the outcome might not be exactly as envisioned.
  • 🎭 Treat the AI as a collaborator, allowing some creativity to be up to the AI for unexpected and sometimes stunning results.
  • ✅ If you're not satisfied with the result, use the remix mode to make adjustments and continue the creative process.

Q & A

  • What is a 'blunt prompt' in the context of creating AI art?

    -A 'blunt prompt' refers to a simple one or two word input, which could be an emotion or an object. The AI then creates an image based on its interpretation of that prompt.

  • How does adding a style to an object in a prompt influence the AI art creation?

    -Adding a style to an object in a prompt allows the AI to not only create an image of the object but also to apply a specific artistic style to it, such as 'DMT style', resulting in a unique and often interesting visual representation.

  • What is meant by 'contrasted Styles' in the context of AI art prompts?

    -Contrasted Styles involve combining two conflicting or contrasting styles to create a unique blend. For example, pairing a traditional style like 'ukiyo-e' with a modern one like '3D render' to give a contemporary twist to a classic art form.

  • Can you explain the concept of 'style stacking' in AI art creation?

    -Style stacking is a method where you take a simple prompt and then continuously add multiple styles on top of it. The AI blends these styles together, allowing for the creation of intricate and layered images. It's akin to building on a base prompt with additional stylistic elements.

  • How does 'chain prompting' work in the process of generating AI art?

    -Chain prompting is a technique where an initial image is created and then used as a reference for subsequent prompts. By copying the link of the generated image and pasting it into new prompts with additional elements, a chain of images is created, each building upon the previous one.

  • What is the significance of 'detailed prompting' in controlling the output of AI art?

    -Detailed prompting involves providing the AI with comprehensive instructions about the main subject, details of the subject, background, art style, lighting, colors, and parameters. This allows for a higher degree of control over the final image, making the outcome more predictable and closer to the creator's vision.

  • How can the aspect ratio and stylization parameters be adjusted in a detailed prompt?

    -In a detailed prompt, you can specify the aspect ratio, such as 16:9, and the level of stylization, indicated by a numerical value like 300. These parameters help refine the output to match the desired composition and visual style.

  • What does the speaker mean by 'remix mode' in the context of AI art generation?

    -Remix mode allows users to modify and tweak the AI's generated images further. By typing '/settings turn on remix', users can introduce additional elements or make changes to the image, providing more control over the final result.

  • Why is it important to not expect the exact image you have in your mind when using AI art generation?

    -It's important because AI interpretation can vary and may not align perfectly with human imagination. Embracing the AI's unique take on the prompt can lead to creative and unexpected results, which can be just as valuable as the original vision.

  • What is the role of the AI in the art creation process as described in the transcript?

    -The AI acts as a tool to interpret and visualize the prompts given to it, creating images based on the input it receives. It can blend styles, create detailed scenes, and generate complex visuals, but it also introduces an element of unpredictability and creativity to the art.

  • How can one get more control over the AI's interpretation in creating art?

    -One can get more control by using detailed prompts that specify the main subject, its details, background, art style, lighting, colors, and parameters. This approach allows for a more directed outcome, although there is still room for the AI's interpretation and creativity.

Outlines

00:00

🖌️ Creative Prompting Techniques for Image Generation

The paragraph introduces various methods to enhance image creation using AI. It begins with the concept of 'blunt prompts', which are simple one or two-word prompts that can be an emotion or an object. The speaker then moves on to more complex formulas, starting with the 'object and style' prompt, where an object is combined with a distinctive style, such as 'DMT style'. The paragraph also discusses 'contrasted styles', where two conflicting styles are merged, and 'contrasted objects', which involves placing an object in an unusual setting. 'Style stacking' is another method mentioned, where multiple styles are layered on top of a base prompt, allowing for intricate and unique results. The paragraph concludes with 'chain prompting', a technique where an initial image is used to inspire a series of subsequent images by using the previous image as a prompt for the next, creating a chain of related images.

05:01

🎨 Advanced Control Over Image Creation with Detailed Prompting

This paragraph delves into a more controlled approach to image generation through detailed prompting. It emphasizes the importance of specifying the main subject, details of the subject, background, art style, lighting, colors, and parameters to guide the AI's interpretation. An example is provided where the speaker describes creating an image of a 'samurai warrior' with 'red armor' that is 'battle damaged', set on a 'beach during sunset' in '3D render Unreal Engine' style with 'cinematic lighting' and 'golden hour' warm tones. The speaker also mentions the use of 'remix mode' for further tweaking and encourages viewers to embrace the AI's creative output even if it differs from their exact vision. The paragraph concludes with a reminder that while the AI can be directed, there is always an element of unpredictability in the final image, and provides a resource (a cheat sheet) for further assistance with parameters.

Mindmap

Keywords

💡Midjourney

Midjourney refers to a hypothetical AI art creation tool mentioned in the transcript. It is used to generate images based on prompts given by the user. The term is central to the video's theme as it sets the context for discussing various prompt formulas and techniques to create stunning AI art.

💡Prompt

A prompt is a stimulus or input that triggers a response or action. In the context of the video, it is a word, phrase, or detailed description that guides the AI to generate a specific image. Prompts are crucial as they are the primary means of communication between the user and the AI art tool.

💡Blunt Prompt

A blunt prompt is a simple one or two-word input, often an emotion or an object, used to generate an image. It is a straightforward approach that relies on the AI's interpretation to create an image, which can result in a 'pretty cool image' as mentioned in the transcript.

💡Object and Style Prompt Formula

This formula involves combining a simple object with a distinct style to generate an image. For example, pairing a 'mythical beast' with a 'DMT style'. The formula is significant as it introduces a method to create images with a specific artistic twist, thus enhancing the creative process.

💡Contrasted Styles

Contrasted styles involve combining two conflicting or contrasting styles to create a unique image. The transcript mentions an example of 'Yukio A' (a traditional style) with '3D render' (a modern style) to achieve a modern twist on traditional art. This concept is key to the video as it demonstrates how to blend different styles for creative results.

💡Contrasted Object

A contrasted object refers to placing an object in an unexpected or contrasting environment. The script provides the example of 'a fish in the desert'. This concept is important as it illustrates how juxtaposing unrelated elements can lead to intriguing and unique images.

💡Style Stacking

Style stacking is a technique where multiple styles are layered on top of a simple prompt to create a complex and detailed image. The transcript mentions an example of 'Pikachu as the Terminator' with additional styles like 'DMT style' or '3D render'. This method is highlighted as a fun and creative way to push the boundaries of AI art generation.

💡Chain Prompting

Chain prompting is a process where an image generated from a prompt is then used as a reference for subsequent prompts, creating a chain of images. The transcript describes using a 'skull and crossbones' image as a starting point and adding elements like 'cyborg in the desert' to continue the chain. This technique is significant as it allows for a continuous evolution of ideas and images.

💡Detailed Prompt

A detailed prompt is a comprehensive description that includes the main subject, details of the subject, background, art style, lighting, colors, and parameters. The transcript provides an example with a 'samurai warrior' in 'red armor' on a 'beach during sunset' with '3D render Unreal Engine' style and 'cinematic lighting'. This approach is important as it allows for greater control and specificity in the image generation process.

💡Remix Mode

Remix mode is a feature that allows users to modify and tweak the generated image further. The transcript instructs users to type '/settings turn on remix' to activate this mode. Remix mode is relevant as it offers additional creative control and the ability to make incremental adjustments to the image.

💡Parameters

Parameters are specific settings or values that can be adjusted to influence the output of the AI art tool. The transcript mentions 'aspect ratio', 'stylize', and provides a link to a 'cheat sheet' for assistance. Parameters are essential as they provide users with fine-tuning options to achieve the desired visual outcome.

Highlights

Using simple one or two word prompts can create cool images based on the AI's interpretation.

Adding a strong style to a simple object, like a mythical beast with DMT style, can produce interesting results.

Contrasted styles involve combining two conflicting styles, such as traditional and modern, to create unique art.

Contrasted objects involve placing an object in an unexpected setting, like a fish in the desert.

Style stacking involves layering multiple styles on top of a simple prompt, allowing for creative combinations.

Chain prompting is a method where you take an existing image and build upon it with additional prompts.

Detailed prompts give more control over the final image by specifying subjects, details, background, style, lighting, colors, and parameters.

Parameters can be adjusted for more control over the image, such as aspect ratio and stylization.

Midjourney AI can be seen as hiring your own artist, providing unique and sometimes unexpected results.

Remix mode allows for further tweaking and customization of the generated images.

Combining prompt formulas can lead to better and more refined results.

The AI's interpretation adds a creative twist, making each image unique.

There is a learning curve to using Midjourney effectively, but the results can be stunning.

The process of creating AI art with Midjourney is iterative, allowing for adjustments and refinements.

Different styles and settings can drastically change the outcome of the generated images.

The AI's interpretation of prompts can lead to surprising and innovative art pieces.

Midjourney can be used for both simple and complex image generation, catering to a wide range of artistic needs.

The video provides a comprehensive guide on how to use Midjourney for creating AI art.