Adaptive Overlapped Block Motion Compensation for Video Encoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video compression methods face challenges with excessive computation and blurring artifacts due to the over-smoothing problem in adaptive overlapped block motion compensation, particularly when handling varying motion between adjacent blocks, which affects video quality and compression efficiency.

Innovation Solution

An adaptive overlapped block motion compensation apparatus and method that selects optimal scan and sampling modes to minimize residual pixel energy and computation cost, allowing for variable unit-based motion compensation that reduces calculation volume and improves video compression performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If adaptive overlapped block motion compensation is performed by pixel unit, then motion reconstruction quality is improved, but computation complexity increases excessively

Engineering Contradiction:
Improvemotion reconstruction qualityVSAvoidcomputation complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the block into multiple sub-blocks (e.g., 4 sub-blocks per block) and applies adaptive OBMC independently to each sub-block. This segmentation allows the system to achieve pixel-level precision where needed while reducing overall computation complexity by processing smaller units independently, thus resolving the contradiction between high reconstruction quality and excessive computation complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different motion compensation strategies to different regions within a block based on local motion characteristics. By determining whether to apply OBMC or conventional BMC separately for each sub-block based on local motion vector differences, the system achieves high reconstruction quality in regions requiring OBMC while avoiding unnecessary computations in regions where conventional BMC suffices, thereby balancing quality and complexity.

Inventive Principle:
Principle #3Local quality

2Object-affected harmful factors

If overlapped block motion compensation is applied to all blocks, then blocking artifacts are reduced, but over-smoothing artifacts occur when adjacent blocks have significantly different motions

Engineering Contradiction:
Improveblocking artifactsVSAvoidover-smoothing artifacts
Core Design Contradiction:
Object-affected harmful factorsVSObject-generated harmful factors

Solution Approach 1:

The patent evaluates motion characteristics locally for each sub-block and only applies OBMC when the motion difference between adjacent sub-blocks is below a threshold. When motion differences are large, the system switches to conventional BMC for those specific sub-blocks, thereby avoiding over-smoothing artifacts while still reducing blocking artifacts in regions where OBMC is appropriate.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent dynamically selects between OBMC and conventional BMC for each sub-block based on real-time motion vector analysis. The decision to apply OBMC or not is made adaptively during encoding based on the specific motion characteristics of each sub-block, allowing the system to respond dynamically to varying motion conditions and avoid over-smoothing while maintaining blocking artifact reduction.

Inventive Principle:
Principle #15Dynamics

3Adaptability or versatility

If fixed unit-based adaptive OBMC is used (block unit, boundary unit, pixel unit), then motion compensation is systematic, but it cannot adapt to varying video characteristics effectively

Engineering Contradiction:
Improveadaptation to video characteristicsVSAvoidsystematic structure
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent divides the block into multiple sub-blocks and applies adaptive OBMC independently to each sub-block based on local motion characteristics. This segmentation provides finer granularity for adaptation compared to fixed block-unit or boundary-unit approaches, enabling better adaptation to varying video characteristics while maintaining a manageable systematic structure through standardized sub-block processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements local adaptive OBMC by evaluating motion characteristics separately for each sub-block and making independent decisions about whether to apply OBMC based on local motion vector differences. This local quality approach provides effective adaptation to varying video characteristics without requiring complex global adaptation mechanisms, thus balancing adaptability with systematic structure.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS8848783B2Motion picture encoding/decoding apparatus, and apparatus and method for adaptive overlapped block motion compensation by variable unit for same
Publication Date: 2014.09.30 SK TELECOM CO LTD
  • US8848783B2 patent drawing
  • US8848783B2 patent drawing
  • US8848783B2 patent drawing

AI summary

The present disclosure relates to a video encoding/decoding apparatus, and apparatus and method for adaptive overlapped block motion compensation by variable units for same. The apparatus and the method for adaptive overlapped block motion compensation by variable units, according to the present disclosure, when conducting a motion compensation, enable an encoder to perform an adaptive overlapped block motion compensation for a plurality of predetermined scan modes and sampling modes, and enable the encoder to calculate the computation volume occurring in said compensation and residual pixel energy, estimate the performance for each mode based on the calculation, determine an optimum scan mode and an optimum sampling mode, thus enabling a decoder to perform a motion compensation with optimum performance and less computing volume based on the determined mode.