Puppet Method:Showing the same characters in multiple scenes

Midjourney allows you to use the "puppet" method, similar to digital animation, where you can stage the same characters repeatedly in different scenes. This technique is beneficial for storytelling and storyboarding purposes.

Simply use this template. You can ignore the details unless you're interested in them. Template http://www.host.com/ReferenceImage.jpg Setting is place:: Name is description:: Setting contains Name doing activity:: atmosphere, filler details, media type, artists --v N --aspect X:Y FOR EXAMPLE… https://as2.ftcdn.net/pFtKgQ.jpg Setting is an empty midnight forest clearing:: Amit is a middle-aged Indian man with short black hair:: Setting contains a frightened Amit wearing pajamas pointing a flashlight into the trees:: ominous rainstorm atmosphere, fairy forest details, photography by Anton Pieck, Frida Kahlo, Beatrix Potter, Gediminas Pranckevicius, Robert Rauschenberg --v 4 --ar 2:3 TLDR: Use this template with the blue keywords. The reference image(s) plus the description of the person (together) act as your reusable Puppet (actor, player, character). You can change the Setting, Activity, Atmosphere, Media Type, etc., around it to stage-direct that character.

Follow the prompt template step-by-step... Method Zero is limited to v4 and involves controlling image-based puppets during the middle of a journey. In v4, you can /imagine with one URL of your character and one URL of your background and do nothing else.Midjourney will use its best judgement to combine the elements, as demonstrated by the combination of a bathroom and a wizard. This approach is effective if you are comfortable with allowing Midjourney to take control.

Method #1: Image-Based Puppets for versions 3 and 4. To have control over the action and scene, it is recommended to include a prompt along with your URLs. Here's a checklist to follow:

- Include reference images with URLs that end in file-format extensions like png, jpg, or gif.

- Provide a short phrase describing the scene, background or setting.

- Add a brief description of your character.

- Mention what the character is doing in another short phrase.

- Consider including style words and/or artist references.

1.To ensure that Midjourney blends your character into the final scene accurately, provide a reference image that closely matches the desired appearance. If you only provide a head-and-shoulders reference, Midjourney may struggle to create a body for the character. Similarly, if the face in the image is too small, it may be difficult for Midjourney to invent a face for it. Therefore, use an uncomplicated or blank background and match the reference image to its ultimate appearance in the final scene.

In version 3 and 4 of Midjourney, multiple URLs can be used. However, in v4 it is recommended to use one picture dedicated to the scene and another devoted to the character as blending effects can quickly become overwhelming otherwise.

Weights are not yet available in v4 but can be used in v3 by adding "::N" after the URL with N being between 0 and 10,000. Experimentation is recommended to find what works best; many people find weights between 0.05 - 6 effective in v3 while some have reported using weights between 0.01 -2 sufficient in v4.

✅Good example of a reference image (load this URL to see the example): https://as2.ftcdn.net/v2/jpg/01/88/09/29/1000_F_188092957_gmFoL2BPkfMfG0h9FzmcgscHRLpFtKgQ.jpg

🚫Bad example of a reference image  (load this URL to see the example) https://as1.ftcdn.net/v2/jpg/04/71/68/56/1000_F_471685627_FfgZX96Z5pa9E3xaOj0AZaCcPDpfnreb.jpg

Please note that Midjourney will interpret the reference image of your scene and character unless it is controlled by the prompt.

2.Please provide a written description of your scene or setting…

To ensure that Midjourney respects your reference image of a scene or setting, make sure that the prompt matches it. If they don't match, you may end up with chaotic blending in your output. However, you can use this effect creatively if desired. For example, I used the same picture of a bathroom but prompted for a jungle to demonstrate this.

When describing your scene or setting, start with "Setting is" (case-sensitive) and keep it brief. The setting should only be the background or stage for the action and not include any players or objects yet. To avoid chaotic blending, make sure that your prompt matches/reiterates any reference images you're using as an anchoring step similar to setting up a background layer.

✅Good: 

Setting is an empty midnight forest clearing::

🚫Bad: 

Setting is an empty forest clearing at midnight where Oberon stands on a tree stump with Puck and a host of fairy soldiers::

Here’s where we are in the v4 wizard/bathroom example:

<https://s.mj.run/CUB6zznrkZo> <https://s.mj.run/NbRZHucYw8k> Setting is a modern bathroom:: Hinkleberry is a wizard in a yellow hat:: Setting contains Hinkleberry carrying a laundry basket:: morning atmosphere, photographed by Anton Pieck, Frida Kahlo, Beatrix Potter, Gediminas Pranckevicius, Robert Rauschenberg --v 4 --q 2 3.Please provide more context or information for me to revise the sentence. To ensure consistency in your character description, the text should match the reference image you use. If there is a discrepancy between them, it can lead to confusion and inconsistency. In such cases, the image reference will likely take precedence unless you adjust its weight.

When describing your character, start with their name (case-sensitive) followed by their defining traits. Avoid adding unnecessary details or dressing up the character as this step serves as an anchor for Midjourney to use your reference images effectively. Keep it concise and focused on essential characteristics.

✅Good:

Amit is a middle-aged Indian man with short black hair::

🚫Bad:

Amit is a man with blue eyes of Indian descent who has normal short hair and some tattoos and wears business casual like a crisp work shirt and pants and leather shoes and a belt and he’s probably a father of three kids and he’s holding a broadsword::

Here’s where we are in the v4 wizard/bathroom example:

<https://s.mj.run/CUB6zznrkZo> <https://s.mj.run/NbRZHucYw8k> Setting is a modern bathroom:: Hinkleberry is a wizard in a yellow hat:: Setting contains Hinkleberry carrying a laundry basket:: morning atmosphere, photographed by Anton Pieck, Frida Kahlo, Beatrix Potter, Gediminas Pranckevicius, Robert Rauschenberg --v 4 --q 2 4.The character is described as doing something in the text… To ensure clarity, conciseness, and coherence in describing your character's actions in a scene or setting, begin with "Setting contains" (case-sensitive). Limit the description to two or three key elements such as the character's attire, emotion, and activity. For instance, changing clothes or providing props can enhance the portrayal of your character. However, keep in mind that image references may overshadow your prompt unless you use weights to increase its influence. See example for guidance.

✅Good: 

Setting contains a frightened Amit with messy hair, wearing pajamas, pointing a flashlight into the trees::

🚫Bad: 

Setting contains the cinematic wide-angle shot of Amit being horrified at a glimpse of a demonic face in the trees and fumbling in his backpack for his flashlight or a weapon and he realizes he’s wearing pajamas and starts crying::

Here’s where we are in the v4 wizard/bathroom example:

<https://s.mj.run/CUB6zznrkZo> <https://s.mj.run/NbRZHucYw8k> Setting is a modern bathroom:: Hinkleberry is a wizard in a yellow hat:: Setting contains Hinkleberry carrying a laundry basket:: morning atmosphere, photographed by Anton Pieck, Frida Kahlo, Beatrix Potter, Gediminas Pranckevicius, Robert Rauschenberg --v 4 --q 2 5.Use the text styling and sourcing prompt to assist MJ in bringing your vision to life. The text styling/sourcing prompt to help Midjourney render your vision is familiar to you from other prompts. I recommend 1-2 atmosphere words, 1-2 filler details, a media type, and some artists. See corresponding colors below for examples. The image reference styles will likely win over your prompt styles unless you use weights to increase the influence of the prompt.

Here’s where we are in the wizard/bathroom example:

<https://s.mj.run/CUB6zznrkZo> <https://s.mj.run/NbRZHucYw8k> Setting is a modern bathroom:: Hinkleberry is a wizard in a yellow hat:: Setting contains Hinkleberry carrying a laundry basket:: morning atmosphere, photographed by Anton Pieck, Frida Kahlo, Beatrix Potter, Gediminas Pranckevicius, Robert Rauschenberg --v 4 --q 2

Here is the complete view of how it looks… https://as2.ftcdn.net/pFtKgQ.jpg Setting is an empty midnight forest clearing:: Amit is a middle-aged Indian man with short black hair:: Setting contains a frightened Amit wearing pajamas pointing a flashlight into the trees:: ominous rainstorm atmosphere, fairy forest details, photography by Anton Pieck, Frida Kahlo, Beatrix Potter, Gediminas Pranckevicius, Robert Rauschenberg --v 3 --ar 9:16 TLDR: Use this template with the blue keywords. The reference image(s) plus the description of the person (together) act as your reusable Puppet (actor, player, character). You can change the Setting, Activity, Atmosphere, Media Type, etc., around it to stage-direct that character. Example v4 wizard/bathroom output…

<https://s.mj.run/CUB6zznrkZo> <https://s.mj.run/NbRZHucYw8k> Setting is a modern bathroom:: Hinkleberry is a wizard in a yellow hat:: Setting contains Hinkleberry carrying a laundry basket:: morning atmosphere, photographed by Anton Pieck, Frida Kahlo, Beatrix Potter, Gediminas Pranckevicius, Robert Rauschenberg --v 4 --q 2

Here is the output for Example v3, which includes the first roll of an Image-Based Puppet with one character reference image.

Method #2: Name-Based Puppet (primarily for v3 - v4 may be too chaotic)

CHECKLIST:

To use this method, you will need the following:

- A unique and descriptive name for your character.

- One text description of a scene, background or setting.

- One text description of your character. (Do not skip)

- One text description of the character performing an action.

- One styling/sourcing prompt to help MJ render your vision.

In version 3, start prompting "A portrait of Made Up Name" until you get a result that resembles the desired character.

This is how Midjourney associates with that particular made-up name. You can summon up or invoke a similar-looking character by using their corresponding made-up name in future prompts.

It's best if the chosen name encodes some characteristics about the person. For instance, "Athena Proudbody" and "Happy Youngman" are more likely to create consistent characters than "Megan Smith" or "Michael McCullough."

For coherence purposes, it's possible to remix or remaster at the end; however, we recommend starting with version 3 when using this method.

Example:

"A portrait of Hipsterman Greenfellow"

"A portrait of Artemis Darkshadow"

"A portrait of Oldman McSmileface"

Prompt until you find a suitable appearance for your character...

From now on, whenever you want to see someone who looks like Oldman McSmileface roughly, simply call him by his name and provide supporting details. Setting is an empty midnight forest clearing:: Oldman McSmileface is an old man with messy hair:: Setting contains a frightened Oldman McSmileface wearing pajamas pointing a flashlight into the trees:: ominous rainstorm atmosphere, fairy forest details, photography by Anton Pieck, Frida Kahlo, Beatrix Potter, Gediminas Pranckevicius, Robert Rauschenberg --ar 9:16

Here is the output for Example v3, first roll, using Name-Based Puppet without a reference image.

Example of Using Puppets to Tell a Story (Version 3) Keep in mind that you need to make multiple rolls to develop any composition. The following example demonstrates the potential of this "puppet" template with just a few rolls. However, please note that this is only a proof-of-concept. Watch the scene, action, and atmosphere change around the core character.