18529817. TRANSFORM-BASED IMAGE CODING METHOD AND DEVICE THEREFOR simplified abstract (LG Electronics Inc.)

From WikiPatents
Jump to navigation Jump to search

TRANSFORM-BASED IMAGE CODING METHOD AND DEVICE THEREFOR

Organization Name

LG Electronics Inc.

Inventor(s)

Moonmo Koo of Seoul (KR)

Jaehyun Lim of Seoul (KR)

Junghak Nam of Seoul (KR)

Seunghwan Kim of Seoul (KR)

TRANSFORM-BASED IMAGE CODING METHOD AND DEVICE THEREFOR - A simplified explanation of the abstract

This abstract first appeared for US patent application 18529817 titled 'TRANSFORM-BASED IMAGE CODING METHOD AND DEVICE THEREFOR

Simplified Explanation

The image decoding method described in the abstract involves performing inverse first-order transform and inverse non-separable transform on a residual sample, based on transform index and multiple transform selection (MTS) index.

  • Inverse first-order transform and inverse non-separable transform are performed on a residual sample.
  • The inverse non-separable transform is based on a transform index indicating a predetermined transform kernel matrix.
  • The inverse first-order transform is based on a MTS index indicating MTS for horizontal and vertical transform kernels.
  • Syntax element bin string for the transform index is derived based on first or second context information depending on the tree type for a split structure of a target block.

Potential Applications

This technology can be applied in image and video processing, compression algorithms, and multimedia applications.

Problems Solved

This technology helps in efficiently decoding images and videos, improving compression ratios, and enhancing overall image quality.

Benefits

The method allows for accurate and efficient decoding of images, leading to better compression and transmission of visual data.

Potential Commercial Applications

This technology can be utilized in digital cameras, video streaming services, video conferencing platforms, and any application involving image or video processing.

Possible Prior Art

One possible prior art could be the use of various transform techniques in image and video compression algorithms, but the specific combination of inverse first-order transform and inverse non-separable transform as described in the patent application may be novel.

Unanswered Questions

How does this method compare to existing image decoding techniques in terms of efficiency and accuracy?

This article does not provide a direct comparison with existing image decoding techniques, so it is unclear how this method stands out in terms of efficiency and accuracy.

What are the potential limitations or drawbacks of implementing this image decoding method in practical applications?

The article does not address any potential limitations or drawbacks of implementing this image decoding method, leaving room for further exploration into its practical implications.


Original Abstract Submitted

An image decoding method according to the present document comprises a step for performing inverse first-order transform and inverse non-separable transform on a residual sample. The inverse non-separable transform is performed on the basis of a transform index indicating a predetermined transform kernel matrix, the inverse first-order transform is performed on the basis of a multiple transform selection (MTS) index indicating MTS for a horizontal transform kernel and a vertical transform kernel, and a syntax element bin string for the transform index is derived on the basis of first context information when a tree type for a split structure of a target block is not a single tree type and is derived on the basis of second context information when the tree type is the single tree type.