This Midjourney Update is WILD + Pika's New AI Video Feature

Curious Refuge
15 Mar 202419:16

TLDRThis week in AI film news highlights the release of a character consistency feature in Mid Journey, enhancing storytelling in film and advertising. Additionally, 11 Labs introduced a sound effects tool, and Open AI's CTO provided an update on Sora's anticipated market release. The video also discusses the potential of style projection in augmented reality and showcases exceptional AI films, demonstrating the growing capabilities and creative applications of AI in the filmmaking process.

Takeaways

  • 🎬 Mid Journey has released a new character consistency feature, allowing for the easy creation of the same character with the same clothing repeatedly, which is a significant advancement for storytelling in film and advertising projects.
  • 👥 The character consistency feature works best with images created within Mid Journey, and using external images may not yield the same quality of results.
  • 🌟 The new tool enables users to create photorealistic shots with consistent characters in various settings, such as New York City, by using image and character references.
  • 👗 The character weight parameter (CW) can be adjusted on a scale from 0 to 100 to control the variability of a character's clothing and appearance, with 0 allowing for more exploration and 100 ensuring consistency.
  • 🎨 Users can blend consistent characters with consistent styles, which is crucial for maintaining a uniform cinematic look across different shots.
  • 🔊 11 Labs has introduced a sound effects tool that can automatically add sound to footage, enhancing the immersive experience of AI-generated videos.
  • 🎥 Pāb Labs also offers the ability to add sound effects to videos, including previously generated content, by typing in the desired sound effects.
  • 🎧 A beta version of a sound effects tool from 11 Labs is available, which allows users to input prompts and receive generated sound effects.
  • 📢 The game of the week involves a challenge for viewers to identify which monster roaring sound effects came from which tool: 11 Labs, Pāb Labs, or a stock sound effects company.
  • 🚀 Elon Musk announced that GPT-4.5 Turbo will be released in June 2024, and the CTO of Open AI hinted that Sora, another AI tool, will be coming out later this year with audio capabilities and restrictions on reproducing public figures.

Q & A

  • What is the new feature released by Mid Journey that has been highly anticipated?

    -The new feature released by Mid Journey is the character consistency feature, which allows users to easily create the same character with the same clothing repeatedly, significantly improving storytelling and film and advertising projects.

  • How does the character consistency feature work in Mid Journey?

    -The character consistency feature works by allowing users to upload an image of a character created within Mid Journey and use it as a reference. This reference helps generate consistent character appearances across different scenes and prompts.

  • What is the significance of the character weight setting in Mid Journey?

    -The character weight setting, adjustable from 0 to 100, determines the consistency of a character's appearance. A weight of 0 allows for more variability in the character's clothing and style, while a weight of 100 ensures the character wears the exact same clothes in every generation.

  • How can users achieve consistent styles in their AI-generated films using Mid Journey?

    -Users can achieve consistent styles by using a combination of character consistency and style referencing. They can search for a cinematic still with a desired color grade or style, use it as a reference, and then generate content that maintains this stylistic consistency.

  • What sound effects tool did 11 Labs release recently?

    -11 Labs released a tool that allows users to automatically add sound effects to their footage. Users can type in the desired sound effects and the tool generates corresponding audio to enhance their videos.

  • How does the sound effects feature in P Labs differ from the one in 11 Labs?

    -While both P Labs and 11 Labs offer sound effects tools, the one in 11 Labs is in limited release and is currently in beta. It allows users to type in the desired sound effects and generates them, whereas P Labs' feature is more mature and integrated into their platform for both new and existing videos.

  • What was the game presented in the AI film news episode?

    -The game presented was a listening test where the audience had to identify which of the three monster roaring sound effects came from which tool: 11 Labs, P Labs, or a stock sound effects company called Artlist.

  • What is the significance of the research paper titled 'Style Gaussian'?

    -The 'Style Gaussian' research paper introduces a technology that can project a style from an uploaded reference image into a Gaussian Splat. This has implications for augmented reality, potentially allowing users to see the world in different artistic styles through AR devices.

  • What update was given about Sora, the upcoming product from Open AI?

    -The CTO of Open AI hinted that Sora will be released later in the year, possibly within the next few months. Sora will incorporate audio and will not be able to reproduce public figures, which is considered a positive move, especially during an election year.

  • What is the significance of Elon Musk's announcement about Grock being open source?

    -Elon Musk announced that Grock, an AI model developed by X (formerly known as OpenAI), will be open source. This is significant as it positions X's AI model in relation to Open AI's offerings and could influence the ongoing legal disputes between the two entities.

  • Can you name some of the best AI films showcased in the AI film news episode?

    -The AI film news episode highlighted 'The Mermaid' by Zahir Khan, 'Uncanny Harriet' by Uncanny Harry, and 'Escape' by Marco as some of the best AI films. These films demonstrate the potential of AI in creating breathtaking visual effects and innovative storytelling.

Outlines

00:00

🎬 Introducing Midjourney's Character Consistency Feature

The paragraph discusses the release of a new character consistency feature in Midjourney, a significant development for AI filmmaking and advertising projects. It highlights the ease of creating consistent characters with the same clothing and appearance, which was challenging before. The script provides a step-by-step guide on using the feature, emphasizing its effectiveness with images originally created in Midjourney. It also introduces the concept of 'character weight,' a tool for varying the costume of a character, allowing for exploration and consistency in storytelling.

05:00

🎨 Blending Character Consistency with Style

This section delves into the combination of character consistency with stylistic consistency in AI filmmaking. It explains how to maintain a consistent cinematic look by using Midjourney's tools to blend character references with specific styles, such as color grading. The script provides an example of creating a cinematic still with a consistent character and style, and discusses the use of style weight to inherit style from a reference photo. It also mentions the availability of the alpha version of Midjourney to users who have generated over a thousand images and the alternative methods available through Discord.

10:01

🎞️ Sound Effects Tools from P Labs and 11 Labs

The paragraph covers the introduction of sound effects tools by P Labs and 11 Labs, which allow users to add sound to their AI-generated footage. It explains the process of using these tools, including adding sound effects to text prompts and images, and editing previous videos with new sound effects. The script also compares the sound quality and effectiveness of the tools from P Labs and 11 Labs, and invites the audience to share their preferences. Additionally, it mentions a game where listeners have to identify the source of different monster roaring sound effects.

15:02

📰 Updates on AI Filmmaking and Open AI

This section provides updates on various aspects of AI filmmaking and Open AI. It discusses a research paper on style projection, the potential of AI in interactive entertainment and augmented reality, a new podcast episode featuring insights on AI, and a short film blending live-action with AI-generated visuals. The script also covers an interview with the CTO of Open AI about the upcoming release of Sora, the company's AI tool, and other developments such as the release of ChatGPT 4.5 and Elon Musk's announcement about making Grock open source.

🏆 Showcase of AI Films and Community Engagement

The final paragraph highlights some of the best AI films created using the discussed tools and techniques. It praises the films 'The Mermaid' by Zahir Khan, 'Uncanny Harriet' by Uncanny Harry, and 'Escape' by Marco for their impressive visual effects, narrative storytelling, and innovative use of AI. The script also encourages viewers to subscribe to the AI Film News for the latest updates and to engage with the content by liking the video and subscribing on YouTube.

Mindmap

Keywords

💡AI film making

AI film making refers to the utilization of artificial intelligence technologies in the creation and production of films. This includes the use of AI for visual effects, character generation, and even scripting. In the context of the video, AI film making is the central theme, with discussions about various AI tools and their applications in creating cinematic content.

💡Character consistency

Character consistency in AI film making is the ability to maintain a uniform appearance of characters across different scenes or frames. This is crucial for maintaining continuity and suspension of disbelief in the narrative. The video highlights a new feature in Mid Journey that allows users to easily create consistent characters with the same clothing and appearance, which is a significant advancement for AI-assisted storytelling.

💡Discord version

The Discord version refers to the adaptation of AI film making tools for use within the Discord platform. This allows users to interact with AI tools and generate content directly within the chat environment. The video explains how to use certain AI film making features on both the desktop version and the Discord version, showcasing the versatility and accessibility of these tools.

💡Character weight

Character weight is a parameter in AI film making tools that determines the level of consistency in a character's appearance. A higher character weight value, such as 100, ensures that the character's clothing and style remain the same across different prompts, while a lower value, like 0, allows for more variability. This concept is essential for balancing creativity with consistency when generating AI film content.

💡Sound effects tool

The sound effects tool is a feature in AI film making platforms that enables users to add audio elements to their visual content. This enhances the immersive experience of the film by synchronizing visual actions with appropriate sounds. The video discusses the release of such a tool by 11 Labs, which allows for the automatic addition of sound effects to footage.

💡Style blending

Style blending in AI film making involves combining visual styles from different sources to create a cohesive and aesthetically pleasing final product. This can include color grading, cinematic styles, or other visual elements that contribute to the overall look and feel of the film. The video emphasizes the importance of maintaining a consistent cinematic look through style blending.

💡Cinematic still

A cinematic still refers to a single, static image that captures a moment in a film-like manner. It is a visual element that conveys a story or emotion through its composition, lighting, and subject matter. In the context of AI film making, cinematic stills are often generated using AI tools to create visually compelling images that can be used in film or advertising projects.

💡Podcast episode

A podcast episode is a single segment of a digital audio program that is part of a larger series. Podcasts typically cover specific topics or themes and are released on a regular basis. In the video, a podcast episode is mentioned as a resource for learning more about AI operations and representation in AI-generated content.

💡Short film

A short film is a brief motion picture that typically conveys a complete story or concept in a condensed format. Short films are often used as a creative outlet for filmmakers and can be used for experimental, educational, or entertainment purposes. In the context of the video, a short film is mentioned as an example of blending live-action footage with AI-generated visuals, showcasing the potential of AI in the future of cinema.

💡Open AI

Open AI is an artificial intelligence research organization that aims to ensure that artificial general intelligence (AGI) benefits all of humanity. The organization develops and releases AI technologies and conducts research to advance the field. In the video, Open AI is mentioned in relation to its upcoming release, Sora, and other AI initiatives.

💡Grock

Grock, in the context of the video, refers to an open-source AI model developed by Elon Musk's company, Neuralink. The decision to make Grock open-source means that the model's code and training materials will be freely available for others to use, modify, and build upon, potentially accelerating AI research and development.

Highlights

Mid Journey releases a new character consistency feature, allowing for the easy creation of the same character with the same clothes repeatedly.

The character consistency feature is particularly useful for storytelling and film and advertising projects.

The feature works best with images created inside Mid Journey, providing more accurate results.

Users can create consistent characters with different clothing options by adjusting the character weight parameter.

The character weight can be adjusted on a scale from 0 to 100, with 100 being the default for consistent clothing.

Blending consistent characters with consistent styles is crucial for maintaining a cinematic look in AI-generated films.

The alpha version of Mid Journey is now open to users who have generated over a thousand images on the platform.

Discord can be used to access the alpha version of Mid Journey and its features, including character consistency.

11 Labs has introduced a sound effects tool that can be added automatically to footage within their platform.

The sound effects tool allows users to add sound effects to both generated videos and previously generated content.

11 Labs' sound effects tool is currently in limited release but offers promising capabilities for enhancing AI-generated content.

A research paper titled 'Style Gaussian' was released, showcasing the potential for projecting styles into augmented reality.

A new podcast episode features Renard Jenkins discussing AI representation and its impact on society.

A short film called 'Another' by Dave Clark blends live-action footage with AI-generated visuals, pushing the boundaries of cinematic storytelling.

The CTO of Open AI provides an update on Sora, hinting at a potential release within the next few months with audio capabilities.

Chat GPT 4.5 Turbo was leaked, suggesting a possible release in June 2021.

Elon Musk announces that Grock will be open source, indicating a shift in AI model positioning.

A demo on X shows a robot integrated with Open AI, capable of interpreting its surroundings in a somewhat humanlike manner.

AI films of the week showcase the best AI-generated films, including 'The Mermaid', 'Uncanny Harriet', and 'Escape', highlighting the creative uses of AI in filmmaking.