Video decoding method, apparatus, and computer program, and video encoding method

By applying geometric transformations to groups of samples within a picture, the video coding technology addresses limitations in handling diverse texture patterns, enhancing compression efficiency and performance.

JP7883671B2Active Publication Date: 2026-07-01TENCENT AMERICA LLC
View PDF 4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
TENCENT AMERICA LLC
Filing Date
2023-10-05
Publication Date
2026-07-01

AI Technical Summary

Technical Problem

Existing video coding technologies face limitations in efficiently utilizing geometric transformations to optimize video compression, particularly in handling different local texture patterns within an input picture, which can limit overall coding performance.

Method used

Applying geometric transformations, such as horizontal and vertical flipping, and rotations, to groups of samples within a picture, rather than the entire picture, to adjust the orientation of Coding Tree Units (CTUs) or Geometry Transform Units (GTUs), allowing for more flexible and efficient encoding and decoding processes.

Benefits of technology

Enhances video coding efficiency by adapting to different local texture patterns, improving compression performance and reducing data volume without sacrificing quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007883671000001
    Figure 0007883671000001
  • Figure 0007883671000002
    Figure 0007883671000002
  • Figure 0007883671000003
    Figure 0007883671000003
Patent Text Reader

Abstract

A video bitstream including a current picture of the video is received. A first group of samples and a second group of samples within the current picture are determined. A first geometric transformation is determined for the first group of samples within the current picture, and a second geometric transformation is determined for the second group of samples within the current picture. The first geometric transformation is configured to adjust an orientation of the first group of samples within the current picture. The second geometric transformation is configured to adjust an orientation of the second group of samples within the current picture. The picture is reconstructed, where the first group of samples are reconstructed based on the determined first geometric transformation and the second group of samples are reconstructed based on the determined second geometric transformation.
Need to check novelty before this filing date? Find Prior Art