US Patent Application 18193769. METHOD AND APPARATUS WITH IMAGE PROCESSING simplified abstract

From WikiPatents
Jump to navigation Jump to search

METHOD AND APPARATUS WITH IMAGE PROCESSING

Organization Name

Samsung Electronics Co., Ltd.


Inventor(s)

Heesae Lee of Ansan-si (KR)


Kyubeom Han of Daejeon (KR)


Jaeyoon Kim of Daejeon (KR)


Sung-Eui Yoon of Daejeon (KR)


Seeha Lee of Daejeon (KR)


Inwoo Ha of Seongnam-si (KR)


METHOD AND APPARATUS WITH IMAGE PROCESSING - A simplified explanation of the abstract

  • This abstract for appeared for US patent application number 18193769 Titled 'METHOD AND APPARATUS WITH IMAGE PROCESSING'

Simplified Explanation

This abstract describes a method and apparatus for image processing. The method involves determining sample points on a camera ray based on view-generation information, determining the location statuses of these sample points using a virtual cylindrical coordinate system, projecting and rendering a pixel value for the camera ray based on the location statuses, applying the view-generation information to two neural networks (one trained for foreground images and the other for background images) to generate rendering results, and finally blending these results to generate a rendered image.


Original Abstract Submitted

A method and apparatus with image processing are provided. A method includes determining sample points sampled on a camera ray, wherein the camera ray is based on view-generation information comprising a scene viewing-position and a scene-viewing direction, determining location statuses of the respective sample points based on a virtual cylindrical coordinate system defined by a center and a radius, projecting and rendering a value of a pixel corresponding to the camera ray by, according the location statuses, applying the view-generation information to a first neural network to generate a first rendering result and to a second neural network to generate a second rendering result, wherein the first neural network has been trained to generate foreground images and the second neural network has been trained to generate background images, and generating a rendered image based on blending the first rendering result and the second rendering result.