18055594. GENERATING ADAPTIVE THREE-DIMENSIONAL MESHES OF TWO-DIMENSIONAL IMAGES simplified abstract (ADOBE INC.)

From WikiPatents
Jump to navigation Jump to search

GENERATING ADAPTIVE THREE-DIMENSIONAL MESHES OF TWO-DIMENSIONAL IMAGES

Organization Name

ADOBE INC.

Inventor(s)

Matheus Gadelha of San Jose CA (US)

Radomir Mech of Mountain View CA (US)

GENERATING ADAPTIVE THREE-DIMENSIONAL MESHES OF TWO-DIMENSIONAL IMAGES - A simplified explanation of the abstract

This abstract first appeared for US patent application 18055594 titled 'GENERATING ADAPTIVE THREE-DIMENSIONAL MESHES OF TWO-DIMENSIONAL IMAGES

Simplified Explanation

The patent application describes a system for generating three-dimensional meshes representing two-dimensional images for editing purposes. The system utilizes neural networks to determine density values of pixels, sample points in the image, generate a tessellation, estimate camera parameters, and modify the mesh based on the camera parameters.

  • Neural networks used for determining density values and camera parameter estimation
  • Sampling points in the image to generate a tessellation
  • Modifying the mesh based on estimated camera parameters
  • Mapping the mesh to the two-dimensional image for editing
  • Updating the two-dimensional image based on modifications to the mesh

Potential Applications

This technology could be applied in various fields such as:

  • Graphic design
  • Virtual reality
  • Augmented reality
  • Animation

Problems Solved

This technology helps in:

  • Enhancing image editing capabilities
  • Creating realistic 3D representations of 2D images
  • Improving visual effects in media production

Benefits

The benefits of this technology include:

  • Streamlining the editing process
  • Enhancing creativity in design and visualization
  • Increasing efficiency in image manipulation

Potential Commercial Applications

Potential commercial applications of this technology could include:

  • Software development for graphic design and animation
  • Tools for virtual and augmented reality content creation
  • Integration into existing image editing software for enhanced capabilities

Possible Prior Art

One possible prior art could be the use of neural networks in image processing and editing tools. Another could be the use of 3D modeling software for creating meshes from 2D images.

Unanswered Questions

How does the system handle complex images with multiple layers or textures?

The patent application does not provide specific details on how the system handles complex images with multiple layers or textures. Further information on this aspect would be helpful for understanding the system's capabilities in handling intricate images.

What is the computational cost of implementing this system?

The patent application does not mention the computational cost of implementing this system. Understanding the computational requirements would be essential for assessing the feasibility of integrating this technology into existing software or systems.


Original Abstract Submitted

Methods, systems, and non-transitory computer readable storage media are disclosed for generating three-dimensional meshes representing two-dimensional images for editing the two-dimensional images. The disclosed system utilizes a first neural network to determine density values of pixels of a two-dimensional image based on estimated disparity. The disclosed system samples points in the two-dimensional image according to the density values and generates a tessellation based on the sampled points. The disclosed system utilizes a second neural network to estimate camera parameters and modify the three-dimensional mesh based on the estimated camera parameters of the pixels of the two-dimensional image. In one or more additional embodiments, the disclosed system generates a three-dimensional mesh to modify a two-dimensional image according to a displacement input. Specifically, the disclosed system maps the three-dimensional mesh to the two-dimensional image, modifies the three-dimensional mesh in response to a displacement input, and updates the two-dimensional image.