Midjourney V5 - How To Upload A Reference Image Or Art And Use As A Prompt - Detailed Tutorial

Curtis Pyke
16 Mar 202303:09

TLDRIn this tutorial, the presenter demonstrates how to use an image as a prompt to generate art with Midjourney V5. They start by uploading an image to Discord and copying its link. Then, they create a prompt with a description and combine it with the image link using the 'image weight' parameter to control the influence of the original image on the generated art. The process results in highly realistic and detailed images that closely resemble the original, showcasing the impressive capabilities of Midjourney V5.

Takeaways

  • 🖼️ The tutorial is about using Midjourney V5 to upload a reference image and use it as a prompt to create art.
  • 🔍 The process begins by finding an image, like a picture of a lady found on Pixiv.
  • 📁 The image is then uploaded to a Discord server by dragging and dropping it.
  • 🔗 After uploading, the image link is copied for later use in the prompt.
  • 📝 The next step is to create a standard prompt using the command '/imagine' and describing what the image should look like.
  • 🎨 The prompt includes details like depth of field, lens type, lighting, natural features, and photorealism.
  • 🔗 The copied image link is pasted into the prompt using the space bar and then command V or control V.
  • 🔑 An important part of the prompt is the '--IW' parameter, which sets the image weight to influence how much the generated art resembles the original image.
  • 🎚 The image weight can be adjusted from 0.5 (lowest) to 2.0 (highest) to control the influence of the original image.
  • 🖌️ The tutorial demonstrates how to upscale the generated images to improve their quality.
  • 🤩 The result showcases the impressive capabilities of Midjourney V5 in creating realistic and detailed art from a reference image.

Q & A

  • What is the purpose of the tutorial provided in the transcript?

    -The purpose of the tutorial is to show users how to upload a reference image or art and use it as a prompt in the generation process with Midjourney V5 to create characters or art based on the uploaded images.

  • Where did the speaker find the picture of the lady used in the tutorial?

    -The speaker found the picture of the lady on a website called Pixels.

  • What is the first step in uploading an image to Discord according to the tutorial?

    -The first step is to go to your Finder or Explorer, grab the image, and then drag and drop it into Discord.

  • Why is it important to copy the link of the uploaded image in Discord?

    -It is important to copy the link because it will be used later in the prompt engineering process to reference the image.

  • What does the speaker describe as the standard prompt engineering process?

    -The standard prompt engineering process involves using the forward slash imagine command followed by a description of what you want the image to look like.

  • What additional elements does the speaker include in the prompt besides the description of the image?

    -The speaker includes elements such as depth of field (dof), lens type (35 millimeter), lighting (natural), natural features, and photorealism in the prompt.

  • What command is used to insert the copied image link into the prompt?

    -The speaker uses the command 'command V' or 'control V' to paste the copied image link into the prompt.

  • What does 'dash-dash IW' command do in the prompt?

    -The 'dash-dash IW' command sets the image weight, determining how much influence the original image will have on the generated output.

  • What is the range of image weight values mentioned in the tutorial?

    -The range of image weight values mentioned is from 0.5, which is the lowest weight, to 2.0, which is the highest weight.

  • How does the speaker ensure the generated image closely resembles the original image?

    -The speaker sets the image weight to 2.0, which is the highest weight, to ensure the generated image closely resembles the original image.

  • What is the final step in the tutorial after setting the image weight?

    -The final step is to hit enter to generate the image based on the prompt and the uploaded reference image.

  • What does the speaker recommend doing with the generated images?

    -The speaker recommends upscaling the generated images that closely match the desired outcome, specifically mentioning images one and three as examples.

Outlines

00:00

🎨 Transforming an Image into Art with Mid-Journey Version 5

This paragraph introduces a tutorial on how to use the mid-journey version 5 software to transform an image into art. The speaker begins by showing an example image found on pixels and explains the process of using this image as a prompt to generate characters. The speaker then provides examples of the impressive results that can be achieved with this method. The tutorial proceeds with a step-by-step guide on how to upload the image to Discord, copy its link, and use it in conjunction with a descriptive prompt to create a photorealistic image with a depth of field, natural lighting, and natural features. The importance of image weight (IW) is highlighted, with the speaker choosing a high weight of 2.0 to ensure the generated image closely resembles the original.

Mindmap

Keywords

💡Midjourney V5

Midjourney V5 refers to the fifth version of a software or tool, likely related to image generation or editing, given the context of the tutorial. It is the central focus of the video, as the host demonstrates how to use this version to create art from an image. The term is used to describe the software's capabilities and to guide viewers on how to utilize its features effectively.

💡Upload

In the context of this video, 'upload' is the action of transferring a file, such as an image, from a local computer to a server or an online platform. The host instructs viewers to upload an image to a Discord server, which is a necessary step in the process of using the image as a prompt for generating art with Midjourney V5.

💡Reference Image

A 'reference image' is a visual file that serves as a guide or inspiration for creating new content. In this video, the reference image is a picture of a lady that the host found and used to generate new art. It is a key component in the process, as it influences the style and appearance of the generated artwork.

💡Prompt

In the context of AI and image generation, a 'prompt' is a set of instructions or a description that guides the software to create specific content. The host uses a prompt to describe what the generated image should look like, such as 'lady reading a book,' which Midjourney V5 then uses to create the artwork.

💡Discord

Discord is a communication platform that the host uses in the tutorial to demonstrate the uploading process of the reference image. It is a popular choice for gamers and communities, and in this case, it serves as the medium where the image is uploaded and then used for generating art with Midjourney V5.

💡Copy Link

The term 'copy link' refers to the action of duplicating a URL or web address associated with a specific file or webpage. In the tutorial, the host copies the link of the uploaded image, which is a crucial step for incorporating the image into the prompt for Midjourney V5 to generate the desired artwork.

💡Image Weight (IW)

Image weight, abbreviated as 'IW' in the script, is a parameter that determines the influence of the reference image on the generated artwork. The host sets the image weight to 2.0, indicating a high influence of the reference image on the final output, ensuring that the generated art closely resembles the original picture.

💡Upscale

To 'upscale' in this context means to enhance the resolution or quality of an image. The host mentions upscaling the generated images, suggesting that Midjourney V5 provides a feature to improve the clarity and detail of the artwork after it has been initially generated.

💡Photorealistic

Photorealistic refers to the quality of an image or artwork that closely resembles a photograph. In the script, the host includes 'photorealistic' in the prompt to instruct Midjourney V5 to generate an image that looks like a real photograph, adding to the level of detail and realism in the final artwork.

💡Version 5

The term 'version 5' is used to specify the particular iteration of the Midjourney software being used in the tutorial. It is important for viewers to know which version they are working with to ensure they follow the correct steps and achieve the same results as demonstrated in the video.

Highlights

Introduction to a tutorial on using Midjourney V5 to create art from reference images.

Demonstration of turning a found image into art by using it as a prompt in the generation process.

Examples of generated characters from pictures and art, showcasing the capabilities of Midjourney V5.

Step-by-step guide on how to upload an image to Discord and copy its link for use in prompts.

Explanation of standard prompt engineering and the importance of describing the desired outcome.

Inclusion of specific details such as depth of field, lens type, and lighting in the prompt.

Adding the uploaded image link to the prompt using command V or control V.

Use of the '--IW' parameter to set the image weight, influencing the resemblance to the original image.

Adjusting the image weight from the standard 1.0 to 2.0 for a closer match to the original image.

Ensuring the settings are at version five for optimal results with Midjourney V5.

Preview of the generated images after applying the prompt and image weight settings.

Comparison of the original image with the generated images to emphasize the quality of Midjourney V5.

Recommendation to upscale the generated images for higher resolution and detail.

Final thoughts on the impressive capabilities of Midjourney V5 in creating realistic art from prompts.

Thank you message and closing of the tutorial video.