DEEP LEARNING BASED SELECTION OF SAMPLES FOR ADAPTIVE SUPERSAMPLING

Title:

DEEP LEARNING BASED SELECTION OF SAMPLES FOR ADAPTIVE SUPERSAMPLING

Document Type and Number:

Japanese Patent JP2021197136

Kind Code:

A

Abstract:

To provide an apparatus for facilitating deep learning based selection of samples for adaptive supersampling.SOLUTION: In a computing system, a graphics processing unit includes one or more processing elements to: receive training data comprising input tiles and corresponding supersampling values for the input tiles, where each input tile comprises a plurality of pixels, and train, based on the training data, a machine learning model so as to identify a level of supersampling for a rendered tile of pixels.SELECTED DRAWING: Figure 19

Inventors:

DANIEL POHL
CARL MARSHALL
SELVAKUMAR PANNEER

Application Number:

JP2020205543A

Publication Date:

December 27, 2021

Filing Date:

December 11, 2020

Export Citation:

Click for automatic bibliography generation Help

Assignee:

INTEL CORP

International Classes:

G06T15/00; G06N20/00

Attorney, Agent or Firm:

Tadashige Ito
Tadahiko Ito
Osamu Miyazaki
Naoki Fujimura

Previous Patent: Money processing system, storage device, disinfection device, money processing method, and money

Next Patent: Methods, equipment, electronics, storage media and computer programs for training models