The Craziest Faceswap I've Seen Yet / Midjourney's Future & Two New AI Video Platforms!

Theoretically Media
25 Apr 202410:38

TLDRThe video discusses advancements in AI technology, specifically in the areas of face swapping and AI avatars. AI Katana's face swap technology is highlighted for its impressive realism, even during complex actions like eating. The video also explores the future of Midjourney, a 3D world simulator with potential for 360Β° camera control. Synthesia's Express one model is introduced, offering AI avatars with a range of emotions. Midjourney's new 'style random' feature is demonstrated, showcasing its creative and practical applications. Lastly, two AI video generators, Morph Studios and Nim Video, are presented, both offering unique features like lip sync, character consistency, and style customization.

Takeaways

  • πŸ“ˆ The face-swapping technology has significantly advanced, with AI Katana showcasing a highly realistic and convincing example.
  • πŸŽ₯ There is speculation that the face-swapping video is not real-time, but rather a post-processed video, indicating that real-time face-swapping still has some limitations.
  • πŸš€ Midjourney's 12-month roadmap hints at a shift towards 3D, real-time video, and a non-interactive world simulator with an eventual interactive layer.
  • πŸ§‘β€πŸ’Ό Media Molecule co-founder, Alex Evans, has joined Midjourney as a principal research engineer, which could signal a significant boost in 3D capabilities.
  • πŸ” Midjourney's new 'style random' feature randomizes styles, offering a fun and potentially useful tool for generating varied and unique images.
  • πŸ€– Synthesia's new Express one model for AI avatars introduces avatars with emotions, aiming to make them more engaging and realistic.
  • πŸ“š A beginner's course on Midjourney is available for free, featuring a range of instructors including a dedicated section on Midjourney by the speaker.
  • 🌐 Morph Studios, currently in beta, is an AI video generator that allows for character consistency and style customization with an interesting node-based UI.
  • πŸ“Ή Nim Video is another AI video platform in beta that offers features like image to video conversion, video restyling, and layer-based editing.
  • πŸ“ˆ The demand for AI-generated content is growing, with tools becoming more sophisticated and offering greater control over the output.
  • πŸ”— Both Morph Studios and Nim Video are currently offering beta access, indicating that the field is rapidly evolving and new tools are becoming available to creators.

Q & A

  • What is the main topic discussed in the video?

    -The main topic discussed in the video is the advancements in face swapping technology, the future of Midjourney with its 12-month roadmap, and the introduction of two new AI video platforms.

  • Which company is credited with the advanced face swap technology shown in the video?

    -The advanced face swap technology is credited to AI Katana.

  • What is the speculation regarding the real-time capabilities of the face swapping technology demonstrated?

    -There is speculation that the face swapping technology is not running in real-time, but rather a pre-recorded video processed through face swapping software.

  • What is the name of the new AI avatar model from Synthesia that has emotions?

    -The new AI avatar model from Synthesia is called Expressive AI.

  • What is the expected direction for Midjourney in the next 12 months?

    -The expected direction for Midjourney is to focus on video, 3D, real-time, and creating a non-interactive world simulator with an added interaction layer.

  • Who is the co-founder of Media Molecule that has joined Midjourney?

    -Alex Evans, one of the co-founders of Media Molecule, has joined Midjourney as a principal research engineer.

  • What is the new feature released by Midjourney called?

    -The new feature released by Midjourney is called 'Style Random'.

  • What is the purpose of the 'Style Random' feature in Midjourney?

    -The 'Style Random' feature in Midjourney is used to randomize the style of generated images, allowing for a wide range of stylistic outcomes and creative exploration.

  • What are the two new AI video generators mentioned in the video?

    -The two new AI video generators mentioned are Morph Studios and Nim Video.

  • What is the unique feature of Morph Studios' user interface?

    -Morph Studios' user interface features a node-based structure that allows for the rerolling of different styles and the connection of chosen aspects to the next shot or node.

  • What is the significance of the Orb as mentioned in the video?

    -The Orb is a device speculated to generate and manage thousands of 3D rooms, indicating Midjourney's serious intentions towards 3D capabilities and data collection.

  • How does the 'Style Random' feature in Midjourney become useful?

    -The 'Style Random' feature becomes useful when a user stumbles across a style they like, as they can then continue to use that style for subsequent image generations.

Outlines

00:00

😲 Advanced Face Swapping and AI Avatars

The video discusses the significant advancements in face swapping and AI avatars. It features a demonstration from AI Katana, highlighting the impressive tracking and realism of the technology, even during complex actions like eating. The video also explores the future roadmap of Mid Journey, a platform for content creation, which is focusing on integrating video, 3D, and real-time capabilities. The host expresses skepticism about the real-time capabilities of current face swapping technology but is excited about the potential of AI avatars from Synthesia that can express emotions. The segment ends with a mention of Elon Musk's involvement and a playful comment about Amazon wigs.

05:01

πŸš€ Mid Journey's 12-Month Roadmap and New Features

The host delves into Mid Journey's plans for the next year, which include a shift towards 3D scene generation with full camera control. There is speculation about the 'orb,' a device for managing 3D rooms, and the hiring of Ahmad, a key figure behind the Apple Pencil. The video also covers the new 'style random' feature in Mid Journey, which allows for randomized styles in image generation, leading to diverse and creative results. The host shares a personal anecdote about conducting a Mid Journey course and provides a link for interested viewers.

10:02

🎬 New AI Video Generators: Morph Studios and Nim Video

The video introduces two new AI video generators: Morph Studios and Nim Video. Morph Studios is in beta and offers a node-based UI for creating animated-style videos with lip sync and sound features. The host expresses curiosity about the tool's interface and workflow. Nim Video is also in beta and offers a range of features, including style and character customization, camera motion, lip syncing, and the ability to work in layers. The host mentions Nvidia's use of open-source models and provides a link for viewers to sign up for the beta.

Mindmap

Keywords

πŸ’‘Face Swapping

Face swapping is a technology that involves replacing one person's face in a video with another person's face. It is often used in entertainment and has become more sophisticated with advancements in AI. In the video, it is mentioned as a technology that has taken a significant leap, with AI Katana showcasing a highly convincing face swap that tracks the subject's movements and facial expressions accurately, even during actions like eating.

πŸ’‘AI Avatars

AI avatars are digital representations of a person that can be controlled or directed by AI algorithms. They are used in various applications, from virtual assistants to video games. The video discusses the next generation of AI avatars from Synthesia, which are capable of displaying emotions, making them more engaging and realistic.

πŸ’‘Midjourney

Midjourney is a term used in the video to refer to a 12-month roadmap of a company or project's future developments. It is mentioned in the context of a company that is planning to focus on video, 3D, and real-time technologies to create non-interactive world simulators. The video suggests that these advancements will lead to more immersive and interactive experiences.

πŸ’‘3D Real Time

3D real time refers to the generation of three-dimensional scenes or environments in real time, which can be manipulated and viewed from different angles. The video discusses how this technology might be integrated into Midjourney, allowing for the creation of 360Β° rotational camera control over generated scenes.

πŸ’‘Deepfake

Deepfake is a term used to describe AI-generated synthetic media where a person's likeness is superimposed onto another's body in a video. The video mentions deepfake in the context of the advanced face swapping technology, noting that while it looks good, there are still inconsistencies that can be detected upon close examination.

πŸ’‘Synthesia

Synthesia is a company that specializes in creating AI avatars. The video highlights their new Express one model, which is capable of expressing emotions. This advancement is significant as it allows for more realistic and engaging virtual interactions.

πŸ’‘Morph Studios

Morph Studios is an AI video generator mentioned in the video. It is in beta and offers features such as animated looks, character image uploads for consistent characters and styles, lip sync, and sound. The interface is described as having a node-based structure, allowing for a unique workflow in creating videos.

πŸ’‘Nim Video

Nim Video is another AI video generator in beta, offering a workspace with options for style and character, consistent characters, camera motion, sound, and lip sync. It also includes features like image to video conversion, video restyling, upscaling, and layer-based editing.

πŸ’‘Style Random

Style Random is a feature released by Midjourney that randomizes the style of generated images. The video explains how this feature can be both fun and useful, allowing users to experiment with different styles and then apply a preferred style to new prompts for a consistent aesthetic.

πŸ’‘Media Molecule

Media Molecule is a developer known for creating the 3D creation engine 'Dreams' for the PlayStation. Alex Evans, one of the co-founders of Media Molecule, has joined Midjourney as a principal research engineer, which is significant as it suggests that the company is serious about its 3D development plans.

πŸ’‘Orb

The Orb is described as a device that could generate and manage thousands of 3D rooms. It is mentioned in the video as part of Midjourney's plans, indicating that the company is exploring ways to create and manage complex 3D environments.

Highlights

AI face swapping technology has significantly advanced, with AI Katana showcasing a highly realistic face swap video.

The face swap technology convincingly tracks facial movements, even during eating and touching the face.

Speculation that the video was not generated in real-time but rather processed through software post-capture.

A comparison is made between the current face swapping technology and previous generations like Face Fusion and RoF.

Synthesia introduces Express One, an AI model with pre-trained avatars that can express emotions.

The new AI avatars from Synthesia are more emotive and align precisely with speech.

Midjourney's 12-month roadmap hints at a shift towards 3D, real-time video, and interactive world simulation.

David Holt, CEO of Midjourney, suggests a future where users can control a 360Β° camera within generated scenes.

The hiring of Alex Evans from Media Molecule, known for developing the 3D creation engine 'Dreams', indicates a strong push towards 3D capabilities.

Midjourney's 'Orb' device is rumored to manage thousands of 3D rooms, with Ahmad, a key figure behind Apple's M1 Pro, now leading hardware at Midjourney.

Midjourney's new 'Style Random' feature randomizes styles, offering a fun and useful tool for creative exploration.

The 'Style Random' feature allows users to stumble upon new styles and apply them to future prompts.

Morph Studios, currently in beta, offers a node-based UI for creating animated-style AI videos with lip sync and sound features.

Nim Video, another AI video generator in beta, provides options for style, character, camera motion, and sound, with features like image to video conversion and upscaling.

Nvidia's platform will utilize open-source models, allowing for a broader range of creative possibilities.

The host offers a beginner's course on Midjourney for those interested in getting started with the platform.

The video concludes with a teaser for future in-depth exploration of Morph Studios and Nim Video once access is granted.