Midjourney PROMPT TIPS for more precise control.

Wade McMaster - Creator Impact
4 Oct 202313:18

TLDRThe video provides tips for taking more control over prompts in mid-journey, a creative AI tool. It demonstrates how to adjust prompts for more precise image generation by using the seed parameter for testing and consistency. The video shows how to separate words with colons and assign strength values to emphasize certain elements in the image. It also explains how to use image weights and negative words in prompts for better control. Additionally, the video highlights the impact of commas in prompts and the potential for unique results when using different languages. Lastly, it suggests using permutations to test different phrases and working on prompt formulas for improved results.

Takeaways

  • πŸ”„ Use the seed parameter to recreate similar images by copying the seed number from the bot's direct message and including it in your prompt with `--seed`.
  • πŸ“ Adjust the aspect ratio with `aspect ratio 16:9` for better suitability in certain contexts, like videos.
  • βž— Separate words using double colons (`::`) to prevent the AI from interpreting them as a single entity.
  • πŸ“ˆ Assign strength values to words to control their importance in the generated image, using a colon and a number (e.g., `::3`).
  • πŸ”„ Make subtle changes to prompts to direct the AI's interpretation, such as adding or removing commas for different results.
  • 🚫 Use negative words with `--no` to exclude certain elements from the generated image.
  • πŸ”— Test different languages for unique results by translating the prompt, as the AI may interpret different languages in varied ways.
  • πŸ”„ Utilize image weights with `--image weight` to influence the importance of different images in a multi-image prompt.
  • βž• Combine image prompts with weights to achieve a balance between elements, scoring images directly if necessary.
  • πŸ”§ Experiment with prompt formulas and permutations in curly brackets `{}` to discover new styles and refine your desired outcome.
  • πŸ“ Keep refining your prompts and testing different variations to better communicate with the AI and achieve the desired results.

Q & A

  • What does the seed parameter in mid-journey allow you to do?

    -The seed parameter in mid-journey allows you to recreate the same image by using the same starting point that the AI uses to generate images.

  • How can you separate words in your prompt to influence the image generation differently?

    -You can separate words in your prompt by using double colons (::), which tells the AI to treat the separated words as distinct elements rather than a single object.

  • How do you give more importance to a specific word in your prompt?

    -You can give more importance to a specific word by using double colons (::) followed by a number, which acts as a multiplier to increase the influence of that word on the image generation.

  • What is the purpose of adjusting the aspect ratio in your prompt?

    -Adjusting the aspect ratio in your prompt helps to change the shape of the generated image, making it more suitable for different types of outputs, such as videos.

  • How can you use image weight to influence the prominence of elements in the generated image?

    -You can use the '--IW' or '--image-weight' parameter followed by a number to score the importance of images in your prompt, which influences how prominently those elements appear in the generated image.

  • Does the use of commas in your prompt affect the generated image?

    -Yes, commas in your prompt can affect the generated image by altering the emphasis or focus of the elements within the prompt.

  • How can you exclude certain elements from your generated image?

    -You can exclude certain elements by using the '--no' prefix followed by the element you wish to exclude, which tells the AI to omit that element from the image.

  • What is the benefit of using permutations in your prompt?

    -Using permutations allows you to test different phrases or styles within the same prompt, giving you a variety of results to choose from and potentially discovering new styles.

  • How can you use different languages to get unique results in your image generation?

    -Translating your prompt into different languages and using it in the AI can sometimes yield unique results due to the different ways the AI interprets and processes the translated words.

  • What is the importance of testing and tweaking your prompt?

    -Testing and tweaking your prompt is crucial for gaining more precise control over the image generation process, allowing you to achieve the desired outcome more effectively.

  • How can you refine your prompts to better communicate with the AI?

    -You can refine your prompts by using clear and specific language, separating words for distinct elements, adjusting the importance of words or images, and experimenting with different styles and permutations.

  • What is the role of the aspect ratio in the context of video suitability?

    -The aspect ratio, when set to 16:9 as mentioned in the script, is particularly suitable for video formats, ensuring that the generated images will fit well within video content without distortion.

Outlines

00:00

πŸ“ Controlling Image Generation with Prompt Tweaks

The paragraph explains how to take more control over the image generation process using the seed parameter for testing and understanding. It demonstrates the use of the seed number to recreate an image and how to adjust the prompt to communicate more directly with the bot. The video shows how to separate words within a prompt using colons to change the interpretation of the prompt and how to assign strength values to words to influence the image generation. It also covers the aspect ratio adjustment and the use of negative words to exclude certain elements from the image.

05:02

πŸ” Prioritizing Elements in Image Prompts

This section delves into prioritizing and de-prioritizing elements within an image prompt to achieve desired results. It discusses adjusting the strength of different components in the prompt to control the prominence of objects in the generated image. The video also explores the use of image weights to focus more on certain aspects of the image. It refutes the notion that commas do not affect the prompt and demonstrates their impact through a comparison of images generated with and without commas. Lastly, it introduces the concept of using permutations to test different phrases and styles within a prompt.

10:03

🌐 Exploring Language and Negative Prompts

The final paragraph discusses the impact of using negative words in prompts to exclude specific elements from the generated image. It also highlights the potential for unique results when using different languages, as demonstrated by translating the prompt into Japanese. The video emphasizes the importance of testing different prompt formulas and methods to better communicate with the language model. It also mentions the use of permutations to explore various phrases and styles, and it encourages viewers to work on their prompt formulas for more effective image generation. The speaker shares additional resources, including a video on prompt formulas and a Mid-Journey cheat sheet.

Mindmap

Keywords

Prompt

A prompt is a command or a statement that a user inputs into a system to generate a specific response or action. In the context of the video, it refers to the instructions given to the Midjourney bot to create images. The video discusses how to refine prompts for more precise control over the generated images.

Seed Parameter

The seed parameter is a numerical value used in generating images to ensure consistency and reproducibility. The video explains that by using the same seed number, one can recreate similar images. It is a crucial concept for testing and understanding the impact of different prompts on the generated images.

Aspect Ratio

Aspect ratio refers to the proportional relationship between the width and the height of an image. In the video, the aspect ratio 16:9 is mentioned as a suitable format for videos. It is an important parameter when adjusting prompts to control the dimensions of the generated images.

Colon Separation

Colon separation is a technique used in prompts to separate words and ensure that the bot treats them as distinct elements rather than a single entity. For example, using '::' between 'fire' and 'ant' helps the bot understand that they are separate concepts, not a compound noun.

Strength Value

Strength value is a numerical weight assigned to words in a prompt to indicate their importance or influence on the generated image. By assigning different strength values, one can control the prominence of certain elements within the image. The video demonstrates how to use strength values to emphasize or de-emphasize certain aspects of the prompt.

Image Weight

Image weight is a method of scoring the importance of images used in the prompt. It allows the user to control how much influence an image has on the final generated image. The video shows how to use image weight to focus more on one element over another when combining multiple images in a prompt.

Negative Words

Negative words are terms prefixed with a negative indicator (such as 'no' or '-') to exclude certain elements from the generated image. The video illustrates how to use negative words in prompts to refine the output by removing unwanted features, like the color red from the armor.

Commas

In the context of the video, commas are used to separate elements in a prompt and can affect the interpretation of the prompt by the bot. The video demonstrates that commas do make a difference in how the bot parses and generates images from the prompt, contrary to some earlier beliefs.

Remix Mode

Remix mode is a feature that allows users to modify existing prompts with additional instructions or changes. The video uses remix mode to exclude the color red from an image by adding 'no red' to the existing prompt, showcasing how to fine-tune the generated content.

Permutations

Permutations refer to the different combinations of words or phrases that can be generated from a given set. The video uses curly brackets to create permutations of the phrase 'Golden hour, Sunset, late afternoon' to test various styles and see how the bot interprets these different contexts.

Midjourney Bot

The Midjourney bot is an AI system that generates images based on textual prompts provided by users. The video is centered around providing tips and techniques for more effectively communicating with this bot to achieve desired image outcomes through precise control of prompts.

Highlights

Midjourney allows for more precise control over prompts using various techniques.

The seed parameter can be used for testing and understanding the generation process.

Recreating the same image with the same prompt is possible by using the seed number.

The suffix '--seed' followed by a number can be used to generate images with consistent starting points.

Aspect ratio can be adjusted with 'aspect ratio 16:9' for better video suitability.

Separating words with double colons can influence how the bot interprets the prompt.

Strength values can be assigned to words to control their importance in the generated image.

Adjusting the strength of words can lead to different visual outcomes in the generated images.

Image prompts can be weighted with 'image weight' to focus more on certain elements.

Commas in prompts can affect the interpretation and the resulting image.

Negative words can be used with '--no' to exclude certain elements from the generated image.

Translating prompts into different languages can yield unique results.

Using permutations with curly brackets allows for testing different phrases and styles.

Working on prompt formulas can improve the language model's understanding and generate better results.

Midjourney cheat sheet and style videos can provide additional tips and methods for effective prompting.