Useful information
Prime News delivers timely, accurate news and insights on global events, politics, business, and technology
Useful information
Prime News delivers timely, accurate news and insights on global events, politics, business, and technology
Join our daily and weekly newsletters to get the latest updates and exclusive content on industry-leading AI coverage. More information
Halfway through the tripthe popular AI imaging startup with more than 21 million users On its Discord server alone, it is branching out from AI image creation and editing.
Max Kreminski, leader of Midjourney’s Storytelling Lab, demonstrated the new tool, called “Patchwork”, on a live screen share on Discord and X via Relay.
He clarified that it would be a standalone app that would require Midjourney accounts to log in and that the URL would be available as a “research preview” in the “updates” channel of the Midjourney Discord server. Users will need to connect their Midjourney Discord account to their Google account to access Patchwork Research Preview. The company posted instructions to do so on your X account.
The tool appears to be an infinite, blank, web-based canvas with a “toolbox” on the left side of the browser screen, displaying a variety of buttons labeled for “character”, “event”, ” faction”, “place”, “prop” and “random”, as well as tools such as “note”, “image”, “portal”, “save” and “share”. “Save” downloads a JSON file with links to all the Midjourney images created on the canvas. Midjourney considers each canvas as a separate digital “world.”
To switch between worlds, the user creates a “portal”, a small black circular button.
To generate a new world, the user enters a text message into an edit bar at the top of the “create” screen and selects one or more from a set of 10 different image styles.
This then produces a new whiteboard with a bunch of new still image resources and text boxes or entities known as “snippets”, including input boxes that allow the user to request new images or settings that fit the initial world description, even a completely new AI generated. character descriptions.
In the demo livestream, the character’s name was auto-completed with Marcus “Dizzy” Gillespie, echoing the name of the famous jazz musician. Dragging the description to a new character image creation box produces four new AI-generated images.
When adding new character charts, the user can ask you to create names and characteristics, as well as motivations that can lead to conflict based on a story.
The user can then link characters with lines denoting connections between them. They can also write action sequences and scene descriptions, each of which tells a story. Each character can be used in multiple images and these images can be brought together with a single option.
The user can “share” the board with other Midjourney users who can collaborate, supposedly in real time, with multiple cursors moving across the same shared canvas. According to Kreminski, a single world can support dozens, and even up to 100 users. However, he noted that the more users, the more chaotic the experience would be.
Kreminski said that only logged in users can see the dashboards (for now), but in the future, non-users will be able to see the dashboards. He mentioned that tabletop RPG groups were already using the feature to plot their campaigns.
He also said that version 7 (V7) of Midjourney would include a setting to allow consistency of multiple characters in new and different images.
Kreminski further revealed that there were at least 3 different large language models powering the app, including a streamlined open source one unique to Midjourney.
Ultimately, it appears to be a novel, complex, powerful, somewhat overwhelming but compelling tool for storyboarding. I could easily see it being used by film writers and directors, game designers, comic book creators, and even live theater directors and writers.
In the long term, Kreminski said there was a “very clear path in terms of intensifying the details and interactions in the worlds,” including fully immersive 3D virtual reality scenes, but that would likely be years away.
The news comes as other AI researchers, startups like Fei-Fei Li World laboratoriesand big tech companies like Google They seek to develop AI that can create navigable, immersive 3D worlds online from simple prompts or images.
Additionally, Midjourney creator David Holz joined the announcement livestream to state that the startup would launch multiple model customization modes in the coming days.
Currently, Midjourney allows users to rate images to customize the types of images they want to see in generations and adjust the model to their personal preferences. Now, the startup will allow users to have multiple customized versions that they can switch between.
Additionally, Holz shared that Midjourney would allow users to upload and reference multiple images on boards to guide generations.
Additionally, sometime after Christmas (December 25), Midjourney will introduce video models and a Midjourney V7 AI imager that will offer greater immediate insight.
Holz further revealed that Midjourney is working on three or four new hardware projects and said the startup was “trying to diversify and become a full-fledged research lab… it may take us six months to announce all six things.”