Hash-Based Encoder Decisions for Block Matching and Scene Changes

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video encoding techniques face challenges in efficiently handling large numbers of blocks for matching, motion vector precision decisions, and scene change detection, particularly in screen capture content, leading to decreased performance and inaccurate scene change detection.

Innovation Solution

Incorporating hash-based block matching for building hash tables, determining motion vector resolution, and scene change detection to improve encoder-side decisions, including categorizing blocks and using scene change information for selecting reference pictures and adjusting picture quality during encoding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional block matching techniques are used to match blocks in video encoding, then motion compensation can be performed, but the computational complexity increases significantly when a large number of blocks need to be stored or matched

Engineering Contradiction:

Improveblock matching accuracyVSAvoidcomputational complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical block matching operations with a hash-based indexing system. Instead of directly comparing blocks using complex computational operations, the system uses hash functions to generate indices and retrieve blocks from reference picture buffers, significantly reducing computational complexity while maintaining matching accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces hash tables and index structures as intermediary data structures between the block matching process and the reference picture buffers. These intermediaries organize and accelerate the retrieval of candidate blocks, reducing the direct computational burden of matching large numbers of blocks

Inventive Principle:

Principle #24Intermediary (Mediator)

2Manufacturing precision

If motion vector precision is increased to improve encoding accuracy, then better compression performance is achieved, but the bit rate and computational complexity increase

Engineering Contradiction:

Improvemotion vector precisionVSAvoidencoding efficiency

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

The patent implements dynamic motion vector precision adjustment based on content characteristics. The system analyzes reference picture availability and block matching results to adaptively select between integer-pixel and sub-pixel precision modes, optimizing the balance between accuracy and efficiency for different video content regions

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the precision parameter of motion vectors based on encoding conditions. When reference pictures are available and block matching indicates benefit, sub-pixel precision (e.g., 1/4 pixel) is used; otherwise, integer-pixel precision is used, dynamically adjusting the parameter to optimize encoding efficiency

Inventive Principle:

Principle #35Parameter changes

3Reliability

If traditional scene change detection techniques are used, then scene transitions can be identified, but detection accuracy decreases for screen capture content with graphical overlays and transitions

Engineering Contradiction:

Improvescene change detection accuracyVSAvoidcontent type adaptability

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent applies different detection strategies to different regions and types of content. For screen capture content with graphical overlays, the system uses hash-based block matching to detect uniform regions and their transitions, while for natural video content, traditional pixel-based methods are used, making the detection system adaptable to various content types with high accuracy

Inventive Principle:

Principle #3Local quality

4Speed

If all candidate blocks are stored in hash tables for efficient retrieval, then block matching speed improves, but memory consumption increases

Engineering Contradiction:

Improveblock matching speedVSAvoidmemory consumption

Core Design Contradiction:

SpeedVSQuantity of substance

Solution Approach 1:

The patent extracts and stores only the essential information needed for block matching in hash tables - specifically hash indices and reference picture buffer pointers - rather than storing complete block data. This extraction approach enables fast retrieval through hash indexing while minimizing memory consumption by storing only the minimum necessary data structures

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS20260032266A1Hash-based encoder decisions for video coding

Publication Date: 2026.01.29 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Innovations in encoder-side decisions that use the results of hash-based block matching are presented. For example, some of the innovations relate to ways of building hash tables that include some (but not all) uniform blocks. Other innovations relate to ways of determining motion vector resolution based on results of hash-based block matching. Still other innovations relate to scene change detection, including long-term reference picture selection and picture quality determination during encoding.