Fixed-Point Motion Vectors for Lower-Memory Image Encoding and Decoding

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The increasing demand for high-resolution and high-quality images leads to a significant increase in data volume, resulting in higher transmission and storage costs, necessitating improved image encoding/decoding techniques for efficient compression.

Innovation Solution

The method involves clipping and formatting motion vectors to increase bit depth, scaling, and limiting the dynamic range to improve compression efficiency by reducing memory space and computational complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If image resolution and quality are increased, then image quality is improved, but data volume and transmission/storage costs increase

Engineering Contradiction:

Improveimage qualityVSAvoiddata volume

Core Design Contradiction:

Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent applies parameter changes by converting motion vectors from floating-point format to fixed-point format with increased bit depth (e.g., 18 bits). This format transformation reduces the memory space required to store motion vector information while maintaining sufficient precision for high-resolution image encoding, thereby reducing data volume without compromising image quality

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If motion vector precision is increased, then encoding accuracy is improved, but memory space requirements increase

Engineering Contradiction:

Improvemotion vector precisionVSAvoidmemory space

Core Design Contradiction:

Measurement precisionVSVolume of stationary object

Solution Approach 1:

The patent changes the parameter of motion vector representation from floating-point to fixed-point format with specific bit depth (e.g., 18 bits). This provides sufficient precision for motion compensation in high-resolution images while significantly reducing the memory space required compared to floating-point formats

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces bit depth as an additional dimension for optimizing motion vector storage. By adjusting the bit depth parameter, the system achieves an optimal balance between precision and memory efficiency, transforming the trade-off from a two-dimensional problem into a multi-dimensional optimization solution

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Manufacturing precision

If motion vector bit depth is increased, then encoding accuracy is improved, but computational complexity increases

Engineering Contradiction:

Improveencoding accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent changes the numerical format parameter from floating-point to fixed-point with increased bit depth. Fixed-point arithmetic is computationally more efficient than floating-point arithmetic, as it eliminates the need for complex exponentiation and normalization operations, thereby reducing computational complexity while maintaining encoding accuracy

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12375708B2Image encoding/decoding method and apparatus, and recording medium storing bitstream

Publication Date: 2025.07.29 INTELLECTUAL DISCOVERY CO LTD

AI summary

The present specification discloses a method of decoding an image. The method includes obtaining a motion vector of a collocated block included in a reference picture of a current block in a temporal motion buffer; changing a format of the obtained motion vector; and deriving the motion vector, in which the format is changed, into a temporal motion vector of the current block.