Motion Vector Selection for Video Encoding Efficiency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional direct mode motion vector computation is inefficient when there is no continuity of motion between frames, as it relies on storing and scaling the motion vector of the previously encoded frame, limiting encoding efficiency.

Innovation Solution

A picture encoding method that selects a motion vector from a plurality of stored motion vectors, using a motion vector selection step to generate prediction pictures and encode differences, allowing for efficient motion vector selection even without frame continuity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If conventional direct mode motion vector computation is used, then encoding is simplified by storing and scaling the motion vector of the previously encoded frame, but encoding efficiency deteriorates when there is no continuity of motion between frames

Engineering Contradiction:

Improveencoding simplicityVSAvoidencoding efficiency

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The patent applies preliminary action by pre-computing and storing multiple candidate motion vectors (first, second, and third motion vectors) from different reference frames before the actual encoding process. This allows the encoder to select the most appropriate motion vector candidate without performing complex real-time computations, thereby maintaining encoding simplicity while improving encoding efficiency even when motion is not continuous between frames.

Inventive Principle:

Principle #10Preliminary action

2Productivity

If motion vectors from multiple reference frames are stored and selected, then encoding efficiency improves by enabling efficient motion vector selection without frame continuity, but device complexity increases

Engineering Contradiction:

Improveencoding efficiencyVSAvoidmotion vector storage and selection complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the motion vector selection process into distinct segments: a first motion vector from a first reference frame, a second motion vector from a second reference frame, and a third motion vector from a third reference frame. This segmented approach allows the encoder to independently compute and store motion vectors from different temporal references, reducing the complexity of selecting from multiple candidates while improving encoding efficiency through better motion representation.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS9402087B2Picture encoding method, picture decoding method, picture encoding apparatus, picture decoding apparatus, picture encoding program, and picture decoding program

Publication Date: 2016.07.26 NIPPON TELEGRAPH & TELEPHONE CORP

AI summary

In picture encoding that generates prediction pictures from picture information of frames previously encoded and encodes picture information for each area, the encoding efficiency is improved. A plurality of motion vector candidates are stored in advance in a motion vector storage memory (107). A motion vector selection section (109) selects a motion vector to be used from the plurality of motion vectors stored in advance in the motion vector storage memory (107). A prediction picture generation section (104) generates a prediction picture from a reference picture using the selected motion vector. A differential encoding section (103) encodes a differential between the picture information of the present area and the prediction picture. A motion vector designation encoding section (110) encodes information that designates the motion vector selected by the motion vector selection section (109).