18296938. GENERATING VIDEOS USING DIFFUSION MODELS simplified abstract (Google LLC)

From WikiPatents
Revision as of 02:37, 18 October 2024 by Wikipatents (talk | contribs) (Creating a new page)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

GENERATING VIDEOS USING DIFFUSION MODELS

Organization Name

Google LLC

Inventor(s)

Jonathan Ho of New York NY (US)

Tim Salimans of Utrecht (NL)

Alexey Alexeevich Gritsenko of Amsterdam (NL)

William Chan of Toronto (CA)

Mohammad Norouzi of Richmond Hill (CA)

David James Fleet of Toronto (CA)

GENERATING VIDEOS USING DIFFUSION MODELS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18296938 titled 'GENERATING VIDEOS USING DIFFUSION MODELS

Simplified Explanation: The patent application describes methods, systems, and apparatus for generating an output video based on an input by updating an intermediate representation using a diffusion model.

  • The method involves receiving an input and initializing a current intermediate representation.
  • An output video is generated by updating the intermediate representation at each iteration using a diffusion model to process the input and generate a noise output.
  • The current intermediate representation is updated with the noise output at each iteration.

Key Features and Innovation:

  • Utilizes a diffusion model to process the input and generate a noise output for updating the intermediate representation.
  • Updates the intermediate representation at each iteration to generate an output video conditioned on the input.

Potential Applications:

  • Video processing and editing software.
  • Computer-generated imagery (CGI) in movies and animations.
  • Virtual reality (VR) and augmented reality (AR) applications.

Problems Solved:

  • Efficient generation of output videos based on inputs.
  • Enhancing the quality and realism of generated videos.

Benefits:

  • Improved video generation process.
  • Enhanced visual effects in videos.
  • Increased efficiency in video editing.

Commercial Applications: The technology can be used in video editing software, CGI production for films and animations, and VR/AR applications, potentially revolutionizing the way videos are created and edited in various industries.

Questions about the Technology: 1. How does the diffusion model improve the generation of output videos? 2. What are the potential limitations of using this technology in video production?


Original Abstract Submitted

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating an output video conditioned on an input. In one aspect, a method comprises receiving the input; initializing a current intermediate representation; generating an output video by updating the current intermediate representation at each of a plurality of iterations, wherein the updating comprises, at each iteration: processing an intermediate input for the iteration comprising the current intermediate representation using a diffusion model that is configured to process the intermediate input to generate a noise output; and updating the current intermediate representation using the noise output for the iteration.

Retrieved from "http://wikipatents.org/index.php?title=18296938._GENERATING_VIDEOS_USING_DIFFUSION_MODELS_simplified_abstract_(Google_LLC)&oldid=251258"