Context Dependent Merging for Video Encoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video encoding standards, such as the MPEG-4 AVC standard, face inefficiencies in coding video data due to the need for additional side information for joint coding of neighboring blocks in SKIP and DIRECT modes, which can introduce redundancy and loss in rate-distortion efficiency.

Innovation Solution

Implement context-dependent merging rules for partitions using frame partitioning, allowing partitions with implicit coding rules to be merged with others using distinct merging rules different from those used for explicit coding rules, thereby enabling adaptive joint coding within SKIP and DIRECT modes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If joint coding of neighboring blocks is implemented in SKIP and DIRECT modes, then coding efficiency is improved, but additional side information introduces redundancy and rate-distortion loss

Engineering Contradiction:
Improvecoding efficiencyVSAvoidrate-distortion efficiency
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent applies local quality by differentiating the merging behavior based on the coding mode of neighboring blocks. When a SKIP or DIRECT mode block has SKIP/DIRECT mode neighbors, it uses implicit merging without side information. When it has non-SKIP/DIRECT neighbors, it uses explicit merging with side information. This localized adaptation optimizes coding efficiency while minimizing rate-distortion loss.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent introduces dynamic merging rules that adapt to the local context of neighboring blocks. The merging behavior is not fixed but changes dynamically based on the coding modes of adjacent blocks, allowing the system to optimize between implicit and explicit merging strategies in different local regions of the video frame.

Inventive Principle:
Principle #15Dynamics

2Adaptability or versatility

If additional side information is transmitted for merging, then joint coding of neighboring blocks is enabled, but coding penalty increases due to extra data transmission

Engineering Contradiction:
Improvejoint coding capabilityVSAvoiddata transmission volume
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent applies local quality by differentiating the merging behavior based on the coding mode of neighboring blocks. When a SKIP or DIRECT mode block has SKIP/DIRECT mode neighbors, it uses implicit merging without side information. When it has non-SKIP/DIRECT neighbors, it uses explicit merging with side information. This localized adaptation optimizes coding efficiency while minimizing rate-distortion loss.

Inventive Principle:
Principle #3Local quality

3Device complexity

If tree-structured partitioning is used, then frame partitioning is simplified, but joint coding of neighboring blocks with different parents is prevented

Engineering Contradiction:
Improvepartitioning structureVSAvoidjoint coding flexibility
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent applies merging by combining blocks from different parent nodes in the tree-structured partitioning. The merging procedure allows blocks that would normally be in different coding units to be jointly coded, enabling cross-parent joint coding while maintaining the simplicity of tree-structured partitioning for the majority of blocks.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS8750377B2Method and apparatus for context dependent merging for skip-direct modes for video encoding and decoding
Publication Date: 2014.06.10 INTERDIGITAL MADISON PATENT HLDG
  • US8750377B2 patent drawing
  • US8750377B2 patent drawing
  • US8750377B2 patent drawing

AI summary

There are provided methods and apparatus for context dependent merging for SKIP/DIRECT modes for video encoding and decoding. An apparatus includes an encoder for joint coding partitions of an image region using frame partitioning and a partition merging procedure, wherein any given one of the partitions that use a coding mode with implicit coding rules is subject to being merged, with another one of the partitions, using a distinct merging rule different than that used for a coding mode with explicit coding rules.