Apple inc. (20240104686). Low-Latency Video Matting simplified abstract

From WikiPatents
Jump to navigation Jump to search

Low-Latency Video Matting

Organization Name

apple inc.

Inventor(s)

Srinidhi Aravamudhan of San Jose CA (US)

Adrian P. Lindberg of Karlstad CA (US)

Eshan Verma of Hayward CA (US)

Jaya Vijetha Gattupalli of San Jose CA (US)

Mingshan Wang of Mountain View CA (US)

Ranjit Desai of Cupertino CA (US)

Vinay Palakkode of Santa Clara CA (US)

Low-Latency Video Matting - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240104686 titled 'Low-Latency Video Matting

Simplified Explanation

The patent application describes techniques for implementing a low latency, guidance map-free video matting system for extended reality (XR) platforms. The system can generate alpha mattes in near real-time and in a computationally efficient manner, even with low resolution auxiliary inputs. Once generated, the alpha mattes can be used for various graphical processing operations on captured images.

  • The system is designed for use in XR platforms.
  • It can work with low resolution auxiliary inputs, such as binary segmentation masks.
  • The system can generate alpha mattes in near real-time.
  • It can function on a captured image stream alone in a domain-specific setting.
  • Various alpha-aware graphical processing operations can be performed on captured images based on the generated alpha mattes.

Potential Applications

The technology can be applied in various industries such as entertainment, gaming, virtual reality, augmented reality, and video conferencing.

Problems Solved

The system addresses the challenge of generating alpha mattes in real-time and in a computationally efficient manner, especially in XR environments where low latency is crucial.

Benefits

The system provides high-quality visual results, reduces computational costs, and enhances user comfort in XR experiences.

Potential Commercial Applications

The technology can be used in XR content creation tools, video editing software, virtual try-on applications, and interactive XR experiences.

Possible Prior Art

Prior art may include existing video matting systems, real-time image processing techniques, and XR rendering technologies.

Unanswered Questions

1. How does the system handle complex scenes with multiple objects of interest? 2. What are the limitations of the system in terms of input resolution and processing speed?


Original Abstract Submitted

techniques are disclosed herein for implementing a novel, low latency, guidance map-free video matting system, e.g., for use in extended reality (xr) platforms. the techniques may be designed to work with low resolution auxiliary inputs (e.g., binary segmentation masks) and to generate alpha mattes (e.g., alpha mattes configured to segment out any object(s) of interest, such as human hands, from a captured image) in near real-time and in a computationally efficient manner. further, in a domain-specific setting, the system can function on a captured image stream alone, i.e., it would not require any auxiliary inputs, thereby reducing computational costs—without compromising on visual quality and user comfort. once an alpha matte has been generated, various alpha-aware graphical processing operations may be performed on the captured images according to the generated alpha mattes (e.g., background replacement operations, synthetic shallow depth of field (sdof) rendering operations, and/or various xr environment rendering operations).