Motion-Compensated Prediction Images With BIO Gradient Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing moving image coding and decoding techniques, such as those based on bi-prediction motion compensation, lack accuracy in generating prediction images due to insufficient utilization of gradient information.
Innovation Solution
A prediction image generation device that performs motion compensation using multiple reference images, incorporating gradient correction terms derived from multiple gradient images and weight coefficients adjusted based on picture intervals, and employing various filtering and switching mechanisms to enhance prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If bi-prediction motion compensation is used to generate prediction images, then coding efficiency is improved, but prediction accuracy is insufficient due to inadequate utilization of gradient information
Solution Approach 1:
The patent changes the parameters of motion compensation by introducing gradient correction terms and adjusting weight coefficients based on picture intervals. This modifies the prediction model to incorporate gradient information, thereby improving prediction accuracy while maintaining coding efficiency through optimized parameter usage.
Solution Approach 2:
The patent introduces gradient correction terms as intermediary elements that mediate between the reference images and the prediction image. These correction terms, derived from gradient images, serve as a bridge to enhance prediction accuracy by compensating for motion-related errors without requiring additional signaling to the decoder.
2Measurement precision
If gradient correction terms are incorporated into motion compensation, then prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The patent implements a self-service mechanism where the encoder and decoder both perform gradient correction independently using the same algorithm. The encoder generates gradient correction terms from reference images, and the decoder reconstructs them using identical processing, eliminating the need for additional signaling or complex coordination between encoder and decoder.
3Measurement precision
If weight coefficients are adjusted based on picture intervals, then prediction accuracy is enhanced, but computational load increases
Solution Approach 1:
The patent applies partial action by selectively adjusting weight coefficients only for gradient correction terms based on picture intervals, rather than reprocessing the entire prediction image. This targeted approach enhances prediction accuracy where needed while minimizing unnecessary computational operations.
Data Source
Figure 1A~1F
Figure 2A~2H
Figure 3
AI summary
A prediction image generation apparatus for generating a prediction image by performing motion compensation on multiple reference images, the prediction image generation apparatus includes a prediction image generation unit configured to generate the prediction image with reference to a first reference image, a second reference image and a gradient correction term. A motion compensation unit (3091) having a motion compensation function using a bi-prediction gradient change (Bi-directional optical flow, BIO) prediction derives the gradient correction term with reference to each pixel value of a horizontal gradient image and a vertical gradient image. An inter-frame-distance weight derivation unit (309131) included a gradient correction bi-prediction derivation unit (30913) included in the motion compensation unit determines each weight coefficient to be multiplied by the value of the each pixel, depending on a first picture interval and a second picture interval, the first picture interval being a picture interval between a target image to be predicted and the first reference image, the second picture interval being a picture interval between the target image and the second reference image.