Patent · US Active

HDR image representations using neural network mappings

US11361506B2 · kind B2 · utility

1Cited by
9References
12Claims
0Family size

Assignee

Inventors

Key dates

Filing dateApr 8, 2019
Grant dateJun 14, 2022
Priority date
Expiry dateApr 8, 2039

Classification

  • Technology area (CPC H)Electricity
  • CPC primaryH04N19/593
  • WIPO fieldAudio-visual technology
  • WIPO sectorElectrical engineering

Abstract

Methods and systems for mapping images from a first dynamic range to a second dynamic range using a set of reference color-graded images and neural networks are described. Given a first and a second image representing the same scene but at a different dynamic range, a neural network (NN) model is selected from a variety of NN models to determine an output image which approximates the second image based on the first image and the second image. The parameters of the selected NN model are derived according to an optimizing criterion, the first image and the second image, wherein the parameters include node weights and/or node biases for nodes in the layers of the selected NN model. Example HDR to SDR mappings using global-mapping and local-mapping representations are provided.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.