18132272. CONTROLLABLE DYNAMIC APPEARANCE FOR NEURAL 3D PORTRAITS simplified abstract (Adobe Inc.)

From WikiPatents
Jump to navigation Jump to search

CONTROLLABLE DYNAMIC APPEARANCE FOR NEURAL 3D PORTRAITS

Organization Name

Adobe Inc.

Inventor(s)

Zhixin Shu of San Jose CA (US)

Zexiang Xu of San Jose CA (US)

Shahrukh Athar of Stony Brook NY (US)

Sai Bi of San Jose CA (US)

Kalyan Sunkavalli of Saratoga CA (US)

Fujun Luan of San Jose CA (US)

CONTROLLABLE DYNAMIC APPEARANCE FOR NEURAL 3D PORTRAITS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18132272 titled 'CONTROLLABLE DYNAMIC APPEARANCE FOR NEURAL 3D PORTRAITS

The abstract of this patent application describes a method for creating controllable, dynamic appearances for neural 3D portraits by projecting colors and dynamic face normals based on location, surface normal, viewing direction, articulated head pose, and facial expression in a digital video portrait.

  • Project colors at points in a digital video portrait based on location, surface normal, and viewing direction.
  • Project dynamic face normals for points in the digital video portrait based on articulated head pose and facial expression.
  • Disentangle facial appearance in the digital video portrait into intrinsic components in a canonical space based on dynamic face normals.
  • Store and/or render at least a portion of a head pose as a controllable, neural 3D portrait based on the digital video portrait using the intrinsic components.

Potential Applications: - Virtual reality applications - Video game development - Facial recognition technology - Animation and film industry - Medical imaging and analysis

Problems Solved: - Enhancing realism in digital portraits - Providing dynamic and controllable facial expressions - Improving accuracy in facial recognition systems - Enhancing user experience in virtual environments - Streamlining animation and film production processes

Benefits: - Realistic and dynamic 3D portraits - Enhanced user engagement and immersion - Improved accuracy in facial recognition - Streamlined animation and film production - Customizable and controllable facial expressions

Commercial Applications: Title: "Dynamic Neural 3D Portraits: Revolutionizing Virtual Reality and Animation" This technology can be used in virtual reality applications, video game development, facial recognition systems, animation and film production, and medical imaging for enhanced realism and user experience.

Questions about Dynamic Neural 3D Portraits: 1. How does this technology improve facial recognition accuracy in digital portraits? 2. What are the potential implications of this technology for the animation and film industry?

Frequently Updated Research: Stay updated on advancements in neural network technology, facial recognition systems, and virtual reality applications to enhance the capabilities of dynamic neural 3D portraits.


Original Abstract Submitted

Certain aspects and features of this disclosure relate to providing a controllable, dynamic appearance for neural 3D portraits. For example, a method involves projecting a color at points in a digital video portrait based on location, surface normal, and viewing direction for each respective point in a canonical space. The method also involves projecting, using the color, dynamic face normals for the points as changing according to an articulated head pose and facial expression in the digital video portrait. The method further involves disentangling, based on the dynamic face normals, a facial appearance in the digital video portrait into intrinsic components in the canonical space. The method additionally involves storing and/or rendering at least a portion of a head pose as a controllable, neural 3D portrait based on the digital video portrait using the intrinsic components.