Midjourney V5.1 - How To Upload A Reference Image And Use As A Prompt - Detailed Tutorial

Artificial intelligence Prompt
26 May 202304:04

TLDRIn this tutorial, we learn how to upload and use a reference image in Midjourney V5.1 to enhance our image generation process. The process begins by clicking the plus button, then uploading a file from your directory. After selecting the image, copy its address or link. Next, paste this link into the prompt field in Midjourney, followed by your desired prompt. The example used is 'the elastic image woman, bioinformatician working on her computer, that has a 3D projected screen.' Midjourney then generates four variations of the image. To get more variations, use the 'V' option, and to upscale a specific variation, use the 'U' option. The tutorial demonstrates selecting variations and upscaling to achieve a higher quality final image. It concludes with an invitation for viewers to ask questions if they have any.


  • 🔍 **Selecting a Reference Image**: Begin by clicking the plus button and uploading a file from your pictures and folders.
  • 📂 **Choosing the Image**: Choose the specific image you want to use as a reference for your edits in Midjourney.
  • ➡️ **Opening the Image**: After selecting, click 'Open' to proceed with the image upload process.
  • 📌 **Entering the Prompt**: Once the image is uploaded, right-click to copy its address or link.
  • 📋 **Pasting the Link**: Paste the copied link into the command field, followed by a comma and your custom prompt.
  • 💡 **Understanding Variations**: Midjourney provides four variations (V1, V2, V3, V4) for each reference image uploaded.
  • 🔄 **Upscaling Options**: Use 'U1' to upscale a single variation without additional variations.
  • 🔄 **Variation Options**: Use 'V1' to get four more variations of the same peak image.
  • 🖼️ **Viewing Results**: After selecting a variation, you can view the full quality by opening it in a new tab.
  • ⬆️ **Upscaling Final Image**: To upscale the final chosen image, click on the 'YouTube' option for an enhanced result.
  • 📝 **Editing and Queries**: The tutorial assists with editing peaks and encourages users to ask questions if they have any.
  • 📚 **Detailed Tutorial**: The script provides a step-by-step guide on how to upload and use a reference image in Midjourney.

Q & A

  • What is the first step to upload a reference image in Midjourney?

    -The first step is to click on the plus button, then a pop-up will appear where you should click on 'upload a file'.

  • How do you select the image you want to use as a reference in Midjourney?

    -You need to navigate to where your pictures and folders are stored, select the image you want to update, and then click 'open'.

  • What should you do after uploading the image in Midjourney?

    -After uploading, you should click on the picture, right-click on it, and then copy the image address or copy link.

  • How do you paste the copied image link into the prompt in Midjourney?

    -You paste the copied link by clicking in the prompt area and using the keyboard shortcut 'control V' to paste the command link.

  • What does the term 'prompt' refer to in the context of Midjourney?

    -In the context of Midjourney, a 'prompt' is the description you provide to guide the AI in generating the desired output based on the reference image.

  • How many variations does Midjourney typically provide for a given reference image?

    -Midjourney typically provides four variations for each reference image uploaded.

  • What does the 'V' option in Midjourney stand for, and what does it do?

    -The 'V' option stands for 'Variations'. Clicking on 'V1', 'V2', 'V3', or 'V4' will generate four new variations of the corresponding peak.

  • How can you upscale a specific variation in Midjourney?

    -To upscale a specific variation, you click on the 'U' option corresponding to that variation, such as 'U1', 'U2', 'U3', or 'U4'.

  • What happens when you click on 'YouTube' for a peak in Midjourney?

    -Clicking on 'YouTube' for a peak will upscale that peak, providing a higher resolution or more detailed version of the image.

  • Can you get additional variations of an upscaled image in Midjourney?

    -No, the 'U' options are for upscaling a single variation, not for generating additional variations.

  • How can users get help or ask queries about using Midjourney?

    -Users can ask queries by commenting on the tutorial, and the author or community members will reply to help clarify or resolve issues.

  • What is the final step shown in the tutorial for using a reference image in Midjourney?

    -The final step is to click on the upscaled peak to get the final result, which is an enhanced version of the image based on the reference and prompt provided.



📌 Uploading and Editing Reference Images

The video begins with an introduction to the use of reference images and the process of editing them during a project. The host guides viewers through the initial steps, which involve clicking a plus button, uploading a file from the user's directory, and selecting the image they wish to work on. The process continues with entering the image's address or link, and then pasting it into a specific field. The user is then prompted to input a description or 'prompt' for the image, which in this case is an 'elastic image woman, bioinformatician working on her computer with a 3D projected screen.' The video demonstrates how to generate variations of the selected image using the platform's tools, such as V1 to V4 for different variations and U1 to upscale a single image without variations.



💡Reference Image

A reference image is a visual representation used as a guide or inspiration for creating or editing a new image. In the context of the video, it is a crucial part of the process as it helps the user to direct the AI to generate or modify images according to the visual cues provided by the reference image.


Midjourney refers to a specific software or tool that is being used in the video to manipulate and generate images. It is the platform where the user interacts to upload a reference image and apply it as a prompt for further image generation or editing.


To upload means to transfer a file or data from a local system to a remote server or another computer connected via a network. In the video, uploading is the initial step where the user transfers their reference image to the Midjourney platform.


In the context of image generation software, a prompt is a set of instructions or a description that guides the AI to create an image. The video demonstrates how a user can use a reference image and combine it with a textual prompt to direct the AI in generating specific types of images.

💡Elastic Image

The term 'elastic image' in the video refers to a specific output or result generated by the Midjourney tool based on the uploaded reference image and the accompanying prompt. It is a term that might be used to describe the flexibility or adaptability of the generated image to the input provided.


A bioinformatician is a professional who applies computational methods to analyze and interpret biological data. In the video, it is part of the prompt, describing the subject of the desired image as a 'woman bioinformatician working on her computer'.

💡3D Projected Screen

A 3D projected screen is a display technology that projects images into a three-dimensional space, creating a visual effect where the images appear to have depth. In the video, it is mentioned as part of the prompt, indicating that the desired image should include a computer with a 3D projected screen.


Variations refer to different versions or renditions of an image generated by the AI based on the initial input. The video explains that the user can select different variations (V1, V2, V3, V4) to explore diverse outcomes from the same reference image and prompt.


To upscale an image means to increase its resolution or size without losing quality. In the video, clicking on 'U1' is described as a method to upscale a selected variation of the image, enhancing its details and clarity.

💡V1, V2, V3, V4

These abbreviations (V1, V2, V3, V4) represent different variations of the generated image. The video script explains that selecting one of these options will provide the user with a new set of four variations based on the chosen option.

💡U1, U2, U3, U4

Similar to the 'V' series, 'U1, U2, U3, U4' likely represent different levels or stages of upscaling an image. The video indicates that selecting one of these options will upscale the image without providing additional variations.


Starting the process by clicking the plus button

Selecting and uploading a reference image from your files

Choosing the desired image by navigating through folders

Using the 'open' command to load the selected image

Copying the image address or link for further use

Pasting the copied link into the command field

Adding a prompt to the link to guide the image generation

Example prompt: 'The elastic image woman, bioinformatician working on her computer, that has a 3D projected screen'

Receiving and reviewing the generated variations from Midjourney

Selecting a variation for further refinement

Understanding the use of 'V' and 'U' commands for variation and upscaling

Using 'V1', 'V2', 'V3', 'V4' to get additional variations of a selected image

Upscaling a specific variation with the 'U1' command

The result of upscaling an image peak with 'YouTube' command

Final result showcasing an upscaled image with enhanced quality

Guidance offered for users to ask questions and receive replies through comments