Motion Vector Candidate Sorting Using Occurrence Histograms

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video compression systems face challenges in efficiently leveraging motion vector prediction to reduce redundancy in video content, particularly in inter prediction, leading to suboptimal compression efficiency.

Innovation Solution

A method and apparatus for encoding and decoding video that involve obtaining a list of motion vector candidates, creating a histogram of their occurrences, sorting the list based on this histogram, and using selected candidates for encoding or decoding blocks, thereby optimizing motion vector prediction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If motion vector candidates are selected using conventional prediction methods, then encoding complexity is kept low, but compression efficiency is suboptimal

Engineering Contradiction:
Improvecompression efficiencyVSAvoidencoding complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by constructing a histogram of motion vector candidate occurrences before final selection. This preprocessing step analyzes the distribution patterns of available candidates and uses this information to guide the selection process, thereby improving compression efficiency without requiring complex real-time optimization during encoding.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the selection parameter from simple conventional prediction metrics to histogram-based occurrence statistics. By transforming the candidate selection criterion into a probability-based approach where candidates are chosen according to their observed frequency in the histogram, the system achieves better compression while maintaining manageable encoding complexity.

Inventive Principle:
Principle #35Parameter changes

2Loss of information

If motion vector candidates are selected without optimization, then encoding speed is maintained, but redundancy reduction is insufficient

Engineering Contradiction:
Improveredundancy reductionVSAvoidencoding speed
Core Design Contradiction:
Loss of informationVSProductivity

Solution Approach 1:

The patent implements feedback by using the histogram of motion vector candidate occurrences as a statistical model that captures redundancy patterns. This histogram serves as feedback information that guides the selection process, allowing the encoder to exploit observed redundancies and achieve better compression without significantly impacting encoding speed.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

By pre-computing the histogram of motion vector occurrences before final candidate selection, the patent performs preliminary analysis of redundancy patterns. This allows the encoding process to leverage pre-analyzed statistical information for more effective redundancy reduction while maintaining encoding throughput.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4676034A1Occurrence-based MV/BV sorting
Publication Date: 2026.01.07 INTERDIGITAL CE PATENT HOLDINGS SAS
  • EP4676034A1 patent drawingFigure 1A~1B
  • EP4676034A1 patent drawingFigure 1C
  • EP4676034A1 patent drawingFigure 2

AI summary

A method and an apparatus for encoding or decoding an image or a video are provided wherein, for at least one block of the image or video, a list of motion vector candidates for predicting the at least one block is obtained. A histogram of occurrences of the motion vector candidates is determined and the list of motion vector candidates is sorted based on the histogram. The at least one block is encoded or decoded using one or more motion vector candidates selected from the list.