Intel corporation (20240355047). THREE DIMENSIONAL GAUSSIAN SPLATTING INITIALIZATION BASED ON TRAINED NEURAL RADIANCE FIELD REPRESENTATIONS simplified abstract

THREE DIMENSIONAL GAUSSIAN SPLATTING INITIALIZATION BASED ON TRAINED NEURAL RADIANCE FIELD REPRESENTATIONS

Organization Name

intel corporation

Inventor(s)

Alexey M. Supikov of Santa Clara CA (US)

Sainan Liu of San Diego CA (US)

Niloufar Pourian of Los Gatos CA (US)

THREE DIMENSIONAL GAUSSIAN SPLATTING INITIALIZATION BASED ON TRAINED NEURAL RADIANCE FIELD REPRESENTATIONS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240355047 titled 'THREE DIMENSIONAL GAUSSIAN SPLATTING INITIALIZATION BASED ON TRAINED NEURAL RADIANCE FIELD REPRESENTATIONS

The patent application describes a system for implementing three-dimensional Gaussian splatting initialization based on trained neural radiance field representations.

• Apparatus determine a location for an initial three-dimensional (3D) Gaussian splat based on optical densities obtained from a trained neural representation of a scene.
• The optical densities are associated with location sample points along a training ray used to train the neural representation.
• The apparatus set parameters of the initial 3D Gaussian splat based on one of the optical densities associated with the location of the initial 3D Gaussian splat and a color value obtained from the trained neural representation.
• The color value is associated with the location of the initial 3D Gaussian splat, which is used to generate a 3D Gaussian splat representation of the scene.

1. 1. 1. Potential Applications

This technology can be used in computer graphics, virtual reality, augmented reality, and other visualization applications.

1. 1. 1. Problems Solved

This technology addresses the challenge of efficiently initializing three-dimensional Gaussian splats based on trained neural radiance field representations.

1. 1. 1. Benefits

The benefits of this technology include improved accuracy and efficiency in generating 3D Gaussian splat representations of scenes, leading to enhanced visual quality in computer graphics and related applications.

1. 1. 1. Commercial Applications

The commercial applications of this technology include video game development, architectural visualization, medical imaging, and scientific simulations.

1. 1. 1. Questions about Three-Dimensional Gaussian Splatting Initialization
         1. 1. How does three-dimensional Gaussian splatting initialization improve scene representation?

Three-dimensional Gaussian splatting initialization improves scene representation by efficiently setting parameters based on trained neural radiance field representations, leading to more accurate and visually appealing results.

1. 1. 1. 1. 2. What are the key advantages of using neural radiance field representations in Gaussian splatting initialization?

Neural radiance field representations provide a more detailed and realistic representation of scenes, allowing for enhanced visual quality and accuracy in rendering.

Original Abstract Submitted

example systems, apparatus, articles of manufacture, and methods are disclosed to implement three dimensional gaussian splatting initialization based on trained neural radiance field representations. example apparatus disclosed herein determine a location for an initial three-dimensional (3d) gaussian splat based on optical densities obtained from a trained neural representation of a scene, the optical densities associated with location sample points along a training ray used to train the neural representation. disclosed example apparatus also set parameters of the initial 3d gaussian splat based on one of the optical densities associated with the location of the initial 3d gaussian splat and a color value obtained from the trained neural representation, the color value associated with the location of the initial 3d gaussian splat, the initial 3d gaussian splat to be used to generate a 3d gaussian splat representation of the scene.