17987586. METHOD AND APPARATUS FOR THREE-DIMENSIONAL RECONSTRUCTION OF A HUMAN HEAD FOR RENDERING A HUMAN IMAGE simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)

From WikiPatents
Jump to navigation Jump to search

METHOD AND APPARATUS FOR THREE-DIMENSIONAL RECONSTRUCTION OF A HUMAN HEAD FOR RENDERING A HUMAN IMAGE

Organization Name

SAMSUNG ELECTRONICS CO., LTD.

Inventor(s)

Taras Andreevich Khakhulin of Pyatigorsk (RU)

Vanessa Valerievna Sklyarova of Moskva (RU)

Victor Sergeevich Lempitsky of Moskva (RU)

Egor Olegovich Zakharov of Krasnogorsk (RU)

METHOD AND APPARATUS FOR THREE-DIMENSIONAL RECONSTRUCTION OF A HUMAN HEAD FOR RENDERING A HUMAN IMAGE - A simplified explanation of the abstract

This abstract first appeared for US patent application 17987586 titled 'METHOD AND APPARATUS FOR THREE-DIMENSIONAL RECONSTRUCTION OF A HUMAN HEAD FOR RENDERING A HUMAN IMAGE

Simplified Explanation

The patent application describes a method for reconstructing three-dimensional human heads using a single photo. The method involves using a convolutional neural network to encode the source image into a neural texture, estimating face shape, expression, and head pose using a pre-trained system, and generating an initial mesh. The method then predicts a final head mesh based on the initial mesh and neural texture, and renders a realistic human image.

  • The method uses a single photo to reconstruct a three-dimensional human head.
  • It employs a convolutional neural network to encode the source image into a neural texture.
  • A pre-trained system called detailed expression capture and animation (DECA) is used to estimate face shape, expression, and head pose.
  • An initial mesh is generated based on the single source image and a target image.
  • A predicted mesh is created by combining the initial mesh and neural texture.
  • The method allows for animation and realistic rendering of the reconstructed head in various poses.

Potential Applications

  • Virtual reality and augmented reality applications.
  • Video games and character creation.
  • Facial recognition and biometric systems.
  • Medical imaging and analysis.
  • Animation and film production.

Problems Solved

  • Simplifies the process of reconstructing three-dimensional human heads by using a single photo.
  • Enables realistic rendering and animation of reconstructed heads in various poses.
  • Provides a more efficient and accurate method for capturing facial expressions and head poses.

Benefits

  • Saves time and resources by eliminating the need for multiple photos or complex scanning techniques.
  • Allows for the creation of realistic and detailed virtual representations of human heads.
  • Provides a versatile tool for various industries, including entertainment, healthcare, and security.
  • Enhances the accuracy and efficiency of facial recognition and biometric systems.


Original Abstract Submitted

A method of three-dimensional reconstruction of human heads using a single photo in the form of polygonal mesh, with animation and realistic rendering capabilities for novel head poses is provided. The method includes encoding, by using a first convolutional neural network, a single source image into a neural texture; estimating, by a pre-trained detailed expression capture and animation (DECA) system, a face shape, a facial expression, and a head pose by using the single source image and a target image, and providing an initial mesh; providing a predicted mesh of a head mesh based on the initial mesh and the neural texture; rendering a human image by using the predicted mesh.

Retrieved from "http://wikipatents.org/index.php?title=17987586._METHOD_AND_APPARATUS_FOR_THREE-DIMENSIONAL_RECONSTRUCTION_OF_A_HUMAN_HEAD_FOR_RENDERING_A_HUMAN_IMAGE_simplified_abstract_(SAMSUNG_ELECTRONICS_CO.,_LTD.)&oldid=22984"