NEW Midjourney V6 Features! Zoom Out, Pan, Inpainting Tutorial with Examples | Midjourney AI

Woollyfern
27 Jan 202406:18

TLDRIn this video, the host introduces new features for Midjourney V6, including the ability to zoom out, pan, and use inpainting on images. With zoom out, users can extend the content of an image beyond its original boundaries, which is useful for transforming close-up portraits into full-body images or expanding scenes. The pan feature allows for extending the image in specific directions, which is different from zoom out as it moves the subject around the image. Inpainting, or 'vary region,' enables users to modify specific parts of an image with a new or updated prompt. The video also mentions a V6 style tuner that is in development. The host provides tips on using these features effectively and encourages viewers to experiment with style codes available for V5.

Takeaways

  • ๐Ÿ” **Zoom Out Feature**: Allows users to extend the content of an image beyond its original boundaries, useful for changing close-up portraits to full-body images or expanding scenes.
  • ๐ŸŽจ **Remix Mode**: Enables modification of the prompt text while panning or zooming out, offering more creative control over the image variations.
  • ๐Ÿ“ **Custom Zoom**: Provides the option to adjust the aspect ratio of an image while zooming out, with no limit to the number of consecutive zooms.
  • ๐Ÿ“ **Aspect Ratio Adjustment**: Custom Zoom can be used to change the aspect ratio of an image, either by zooming out or without it, to achieve desired dimensions.
  • ๐Ÿ–Œ๏ธ **Inpainting (Vary Region)**: A feature that lets users modify specific parts of an image with a new or updated prompt, enhancing or changing elements within the image.
  • ๐Ÿ“ท **Panning**: Extends the image in specific directions, adding content incrementally and allowing for movement of the subject within the image frame.
  • ๐Ÿงฉ **Content Variation**: Panning in V6 can create variations by moving the subject around without altering the aspect ratio, unlike zoom out.
  • โž• **Adding Text**: Users can insert text or conversation bubbles into images using the Vary Region feature by selecting the area and inputting the text.
  • ๐Ÿ› ๏ธ **Style Tuner**: Although not covered in the video, a V6 style tuner is in development and will be released soon for further customization.
  • ๐Ÿ“š **Free Style Codes**: The presenter offers free style codes for experimentation in V5, with a link provided in the video description.
  • ๐Ÿ”— **Job ID for Existing Images**: To use new features on existing images, one must locate the upscaled image in the gallery, copy its job ID, and use it in Discord commands.
  • ๐Ÿ“ˆ **Successive Zoom Out**: The video demonstrates a zoom out effect using 12 successive zooms, noting the potential for vignetting but offering solutions like adding 'dark corners' to the prompt.

Q & A

  • What new features have been introduced in Midjourney V6?

    -Midjourney V6 introduces the ability to zoom out, pan, and inpaint your images. These features allow users to extend the content of an image beyond its original boundaries, move the subject around within the image, and modify specific parts of the image using a new or updated prompt.

  • How does the zoom out feature work in Midjourney V6?

    -The zoom out feature allows you to extend the content of an image beyond its original boundaries. You can choose to zoom out by a factor of 2 or 1.5, or use a custom zoom value to adjust the aspect ratio while zooming out. The custom zoom feature also allows you to change the prompt text and parameters.

  • What is the purpose of the 'make square' button in the zoom out feature?

    -The 'make square' button adjusts the aspect ratio of the image to make it a square by either expanding it horizontally or vertically. This button is available when the upscaled image has a non-square aspect ratio.

  • How does the pan feature differ from the zoom out feature?

    -While the zoom out feature keeps the subject in the center of the image, the pan feature allows you to move the subject around. Panning adds content in one specific direction at a time, and it does not add pixels to the images like zoom out does.

  • How can you use the inpaint feature to modify an older image?

    -To use the inpaint feature on an older image, you need to log into Midjourney's website, go to your gallery, locate the upscaled image, copy its job ID, and then use that ID in Discord to bring the image back for editing. You can then use the inpaint feature to modify specific parts of the image with a new or updated prompt.

  • What is the 'vary region' feature and how is it used?

    -The 'vary region' feature is Midjourney's inpainting feature that allows you to modify specific parts of an image. You can select an area using the rectangle or lasso tool and then change the prompt text to modify the selected area. It can be used to add objects, remove objects, or add text to the image.

  • How can you ensure better results with the 'vary region' feature?

    -To get better results with 'vary region', it's recommended to select a larger area than a smaller one. If you're not satisfied with the results, try selecting a broader area for the inpaint operation.

  • What is the purpose of adding the '--no' parameter with specific words following it in the prompt?

    -The '--no' parameter followed by specific words will downweight those words in the results, helping to guide the AI to avoid including those elements in the generated image.

  • Can you provide an example of how to use the 'vary region' feature to add text to an image?

    -To add text to an image using 'vary region', you select the area where you want the text to appear and then include the text within quotes in your prompt. For multiple text phrases, you can perform separate 'vary region' operations on the same or an upscaled image.

  • Is there a V6 style tuner available for Midjourney?

    -As of the time of the video, the V6 style tuner is in development and is expected to be released soon.

  • Where can users find free style codes to experiment with in V5?

    -The video description contains a link to free style codes that users can experiment with in Midjourney V5.

  • What should users do if they find the video helpful?

    -If users find the video helpful, they are encouraged to like it, subscribe to the channel, and engage with the content.

Outlines

00:00

๐Ÿ–ผ๏ธ Exploring New V6 Features: Zoom Out, Pan, and Imp Paint

In this paragraph, the speaker introduces the audience to the latest features released by Moury for V6. These features include the ability to zoom out, pan, and imp paint on V6 images. The speaker emphasizes the importance of having remix mode enabled to modify the prompt text and create variations of an image. The zoom out feature is explained as a way to extend the content of an image beyond its original boundaries, which is useful for transitioning from close-up portraits to full-body images, fixing cropping issues, or expanding a scene. The custom zoom option is highlighted as a powerful tool for adjusting the aspect ratio while zooming out. The speaker also discusses the potential vignetting effect after multiple zoom outs and suggests adding parameters like 'V yet black border black corners' to the prompt to mitigate this. The pan feature is introduced as a way to extend the image in specific directions, with the ability to change directions and create variations. The speaker explains that panning moves the subject around while zooming out keeps the subject centered. Lastly, the imp paint feature is introduced, which allows for the modification of specific parts of an image using a new or updated prompt. The speaker demonstrates how to use imp paint to add a hat to a dog in an image and to remove objects from a scene.

05:03

๐Ÿ“ Advanced Prompting Techniques and Adding Text with Very Region

The second paragraph delves into advanced prompting techniques for V6, particularly focusing on the use of the 'D-' parameter to downweight certain words in the results. The speaker shares personal experience, noting an increased reliance on this parameter in V6 compared to V5. The paragraph also covers how to add text to images using the very region feature, with an example of adding text in a conversation bubble. The speaker demonstrates adding text to an image by selecting the region and inputting the text in quotes. For adding multiple text phrases, the speaker suggests performing the task in separate steps. The paragraph concludes with a teaser about an upcoming V6 style tuner and offers free style codes for experimentation in V5, which will be linked in the description. The speaker ends with a call to action, encouraging viewers to like, subscribe, and engage with the content.

Mindmap

Keywords

Zoom Out

Zoom Out is a feature that allows users to extend the content of an image beyond its original boundaries. It is particularly useful for transforming close-up images into full-body shots, fixing cropping issues, or expanding a scene to provide a broader view. In the video, the presenter demonstrates how to use the Zoom Out feature to adjust the aspect ratio and create a wider view of an image.

Pan

Pan is a feature that lets users extend an image in specific directions, which is helpful for moving the subject around within the image or adding content incrementally in one direction. Unlike Zoom Out, which keeps the subject centered, Pan allows for the subject to be repositioned. The script mentions that with Remix mode enabled, users can modify the prompt text while panning to create variations of an image.

Inpainting

Inpainting is a feature that enables the modification of specific parts of an image using a new or updated prompt. It is showcased in the video as a powerful tool for adding or removing elements from an image. For instance, the presenter uses Inpainting to add a hat to a dog in an image and to remove the dog from a scene, leaving just the tropical beach.

Remix Mode

Remix Mode is a setting that, when enabled, allows users to modify the prompt text as they pan or zoom out, creating variations of an image. It is integral to the new features of Midjourney V6, as it provides more flexibility and control over the image manipulation process. The video emphasizes the importance of having Remix Mode enabled to fully leverage the capabilities of the new features.

Aspect Ratio

Aspect Ratio refers to the proportional relationship between the width and the height of an image. In the context of the video, adjusting the aspect ratio is a key part of using the Zoom Out feature. The presenter illustrates how to change an image from a 9 by 6 aspect ratio to a wider 3:2, which can enhance the composition and presentation of the image.

Vignetting

Vignetting is a visual effect where the corners of an image become gradually darker. The video mentions that after several consecutive Zoom Outs, there seems to be a bit of vignetting. This effect can be mitigated by adding parameters to the prompt that specifically address the black border or corners.

Job ID

Job ID is a unique identifier for a specific image processing task. In the video, the presenter explains how to retrieve a Job ID from an upscaled image in the Midjourney gallery to apply new features without starting from scratch. The Job ID is copied and used in Discord to bring the image back for further editing.

Custom Zoom

Custom Zoom is an option that allows users to manually set the zoom level and aspect ratio of an image during the zoom out process. It provides a high degree of control, enabling users to tailor the image to their specific needs. The video demonstrates how Custom Zoom can be used to achieve the desired composition and framing.

Style Tuner

Style Tuner is a feature that is mentioned as being in development and will be released soon. It is implied that this tool will allow users to fine-tune the stylistic elements of their images. Although not fully explained in the video, it is suggested that the Style Tuner will offer further customization options for users working with Midjourney V6.

Chaos

Chaos, in the context of the video, refers to a parameter that can be adjusted to control the level of randomness or variation in the generated image. When panning, users can modify parameters like Chaos to achieve different visual effects, making the feature a versatile tool for image manipulation.

Text Images

Text Images involve adding text to an image, which can be done using the Inpainting feature. The video demonstrates how to select a region of the image and input text to create captions or dialogue. This can be used to add narrative elements or additional information to an image, enhancing its storytelling capabilities.

Highlights

Midjourney AI has released new features for its V6 version, including the ability to zoom out, pan, and inpaint images.

To maximize the use of these features, ensure remix mode is enabled in your settings.

Existing images in the Midjourney gallery can be modified with the new features without starting from scratch.

Zoom out feature extends the content of an image beyond its original boundaries, useful for full body images or establishing shots.

The make square button adjusts the aspect ratio of the image to make it square by expanding it horizontally or vertically.

Custom Zoom allows users to change the aspect ratio and zoom level simultaneously, with no limit to the number of consecutive zoom outs.

Vignetting may occur after several zoom outs, and adding parameters like '--v 5 --ar 3:2' can sometimes help mitigate this.

Pan feature extends an image in specific directions, allowing movement of the subject within the image.

Panning in V6 does not add pixels to the image, unlike zoom out, and can be used to create variations.

Inpainting feature enables modification of specific parts of an image using a new or updated prompt.

For inpainting, selecting a larger area than necessary is recommended for better results.

Inpainting can be used to add objects, such as a hat on a dog, or to remove objects from an image.

Text can be added to images by selecting the region and inputting the desired text within quotes.

Multiple text phrases can be added in separate steps using the inpaint feature.

A V6 style tuner is in development and will be released soon.

Free style codes for experimentation in V5 are available and will be linked in the video description.

The video provides examples and tips for using the new features of Midjourney V6.