18653609. MESSAGING SYSTEM WITH NEURAL HAIR RENDERING simplified abstract (Snap Inc.)

From WikiPatents
Jump to navigation Jump to search

MESSAGING SYSTEM WITH NEURAL HAIR RENDERING

Organization Name

Snap Inc.

Inventor(s)

Artem Bondich of Marina del Rey CA (US)

Menglei Chai of Los Angeles CA (US)

Olekssandr Pyshchenko of Pacific Palisades CA (US)

Jian Ren of Marina Del Ray CA (US)

Sergey Tulyakov of Santa Monica CA (US)

MESSAGING SYSTEM WITH NEURAL HAIR RENDERING - A simplified explanation of the abstract

This abstract first appeared for US patent application 18653609 titled 'MESSAGING SYSTEM WITH NEURAL HAIR RENDERING

Simplified Explanation:

The messaging system described in the patent application uses neural networks to render hair in images uploaded by users.

Key Features and Innovation:

  • Processing 3D models of fake hair and real hair images to generate a fake hair structure
  • Encoding the fake hair structure to generate a coded fake hair structure
  • Processing the coded fake hair structure to generate a fake and real hair structure
  • Encoding a second real hair image to generate an appearance map
  • Synthesizing a real image using the appearance map and the fake and real hair structure

Potential Applications: This technology could be used in virtual reality applications, gaming, and social media platforms to enhance user-generated content with realistic hair rendering.

Problems Solved: This technology addresses the challenge of accurately rendering hair in images, especially in virtual environments where realism is crucial.

Benefits: The technology allows for more realistic and detailed hair rendering in images, enhancing the overall visual quality of user-generated content.

Commercial Applications: The technology could be valuable for companies developing virtual reality experiences, gaming platforms, and social media apps looking to enhance user engagement through realistic visuals.

Prior Art: Researchers in the field of computer graphics and artificial intelligence have explored similar techniques for hair rendering in images.

Frequently Updated Research: Researchers are continually improving neural network algorithms for more accurate and efficient hair rendering in images.

Questions about Hair Rendering Technology: 1. How does this technology improve the realism of hair rendering in images? 2. What are the potential limitations of using neural networks for hair rendering in images?


Original Abstract Submitted

A messaging system performs neural network hair rendering for images provided by users of the messaging system. A method of neural network hair rendering includes processing a three-dimensional (3D) model of fake hair and a first real hair image depicting a first person to generate a fake hair structure, and encoding, using a fake hair encoder neural subnetwork, the fake hair structure to generate a coded fake hair structure. The method further includes processing, using a cross-domain structure embedding neural subnetwork, the coded fake hair structure to generate a fake and real hair structure, and encoding, using an appearance encoder neural subnetwork, a second real hair image depicting a second person having a second head to generate an appearance map. The method further includes processing, using a real appearance renderer neural subnetwork, the appearance map and the fake and real hair structure to generate a synthesized real image.