Bidirectional Optical Flow Motion Compensation Texture Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The bidirectional optical flow (BIO) technique in video encoding and decoding requires high computation complexity, leading to image quality degradation, which needs to be reduced while maintaining effective motion compensation.
Innovation Solution
A method is introduced to selectively apply the BIO process based on texture complexity, block size, motion information encoding mode, constant velocity constraint, brightness constancy constraint, and motion vector variance, using a device with a reference block generator, skip determiner, and prediction block generator to generate prediction blocks by either typical motion compensation or BIO process depending on these conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If bidirectional optical flow (BIO) process is applied for motion compensation, then motion estimation precision is improved, but computation complexity increases
Solution Approach 1:
The patent applies different motion compensation strategies to different regions of the image based on their characteristics. Regions with high texture complexity undergo full BIO processing for accurate motion estimation, while regions with low texture complexity use simplified motion compensation. This local differentiation resolves the contradiction by concentrating computational resources where they provide the most benefit.
Solution Approach 2:
The patent changes the parameter of texture complexity as a criterion to dynamically adjust the motion compensation process. By calculating texture complexity metrics and using them to determine whether to apply BIO, the system adapts the computation level to the local image characteristics, thereby improving motion estimation precision where needed while reducing overall computation complexity.
2Manufacturing precision
If bidirectional optical flow (BIO) process is applied for motion compensation, then image quality is improved, but processing time increases
Solution Approach 1:
The patent applies different motion compensation strategies to different regions of the image based on their characteristics. Regions with high texture complexity undergo full BIO processing for accurate motion estimation, while regions with low texture complexity use simplified motion compensation. This local differentiation resolves the contradiction by concentrating computational resources where they provide the most benefit.
Solution Approach 2:
The patent applies BIO processing partially rather than uniformly across the entire image. By selectively applying BIO only to regions where it provides significant quality improvement (high texture complexity regions), the system achieves adequate image quality without the full processing time cost of universal BIO application.
3Measurement precision
If bidirectional optical flow (BIO) process is applied for motion compensation, then motion vector accuracy is improved, but computational cost increases
Solution Approach 1:
The patent applies different motion compensation strategies to different regions of the image based on their characteristics. Regions with high texture complexity undergo full BIO processing for accurate motion estimation, while regions with low texture complexity use simplified motion compensation. This local differentiation resolves the contradiction by concentrating computational resources where they provide the most benefit.
Solution Approach 2:
The patent changes the parameter of texture complexity as a criterion to dynamically adjust the motion compensation process. By calculating texture complexity metrics and using them to determine whether to apply BIO, the system adapts the computation level to the local image characteristics, thereby improving motion vector accuracy where needed while reducing overall computational cost.
Data Source
AI summary
Disclosed herein is a method for adaptive bidirectional optical flow estimation for inter-screen prediction compensation during video encoding. The method aims to reduce complexity and/or cost of bidirectional optical flow (BIO) at a pixel level or a subblock level.