Microsoft Technology Licensing, LLC

Adobe, Inc.

Mixamo, Inc.

Graham Taylor

Stefano Koenig Corazza

Nazim Kareemi

Edilson De Aguiar

Alejandro Acero

Michael Lewis Seltzer

Systems and methods are described for animating 3D characters using synthetic motion data generated by generative models in response to a high level description of a desired sequence of motion provided by an animator. In a number of embodiments, an animation system is accessible via a server system that utilizes the ability of generative models to generate synthetic motion data across a continuum to enable multiple animators to effectively reuse the same set of previously recorded motion capture data to produce a wide variety of desired animation sequences. One embodiment of the invention includes a server system configured to communicate with a database containing motion data including repeated sequences of motion, where the differences between the repeated sequences of motion are described using at least one high level characteristic. In addition, the server system is configured to train a generative model using the motion data, the server system is configured to generate a user interface that is accessible via a communication network, to receive a high level description of a desired sequence of motion via the user interface, to use the generative model to generate synthetic motion data based on the high level description of the desired sequence of motion and to transmit a stream via the communication network including information that can be used to display a 3D character animated using the synthetic motion data.

Interactive design, synthesis and delivery of 3D motion data through the web

Systems and methods are described for animating 3D characters using synthetic motion data generated by generative models in response to a high level description of a desired sequence of motion provided by an animator. An animation system is accessible via a server system that utilizes the ability of generative models to generate synthetic motion data across a continuum to enable multiple animators to effectively reuse the same set of previously recorded motion capture data to produce a wide variety of desired animation sequences. An animator can upload a custom model of a 3D character and the synthetic motion data generated by the generative model is retargeted to animate the custom 3D character.

Interactive design, synthesis and delivery of 3D character motion data through the web

The quality of sound recorded from a plurality of people speaking at the same time is improved by incorporating prior knowledge into an independent component analysis (ICA) separating algorithm. More particularly, prior knowledge is defined as a probability distribution according to some prior situation (e.g., prior distribution of people in a room). A mixture of sounds (e.g., mixture of voices) from a plurality of sources (e.g., people) captured by one or more recording devices (e.g., microphones) is separated into individual components (e.g., individual voices from respective people) by applying an maximum a posteriori (MAP) ICA algorithm which incorporates prior knowledge of the respective sources (e.g., location of sources) directly into the MAP ICA algorithm thereby allowing recovery of independent underlying sounds associated with individual sources from the mixture. Therefore, incorporating prior knowledge into an ICA algorithm provides sound quality substantially equal to existing ICA systems, but at reduced computational complexity.

Incorporating prior knowledge into independent component analysis

Growing community of inventors

Toronto, Canada

Graham Taylor