Geometrically Modified Reference Pictures for Video Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image compression technologies face challenges in maintaining prediction efficiency due to global motion in reference pictures and restricted intra prediction directions, leading to degraded image quality and compression performance.
Innovation Solution
The method involves generating a geometrically modified picture by altering a reference picture, performing inter prediction using this modified picture, and selecting the final prediction block with fewer errors, along with encoding and decoding geometric modification information to enhance prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If inter prediction is performed using a reference picture with global motion, then the reference picture can be utilized for prediction, but the similarity between the reference picture and current picture is lowered, degrading prediction efficiency
Solution Approach 1:
The patent applies preliminary action by generating a geometrically modified reference picture before performing inter prediction. The encoder/decoder creates a corrected reference picture that compensates for global motion (such as zooming, panning, or rotating) that occurred between the reference picture and current picture. This preliminary geometric correction ensures that when inter prediction is subsequently performed, the reference picture and current picture have higher similarity, thereby improving prediction efficiency without sacrificing the ability to utilize reference pictures with global motion.
2Device complexity
If intra prediction direction is restricted, then the prediction process is simplified, but the efficiency of intra prediction is deteriorated
Solution Approach 1:
The patent applies dynamics by making the intra prediction direction adaptive rather than fixed or restricted. The encoder/decoder determines the intra prediction direction dynamically based on the characteristics of the current block and surrounding blocks. This allows the prediction process to adapt to different scenarios, improving intra prediction efficiency while maintaining reasonable complexity through intelligent direction selection rather than exhaustive search.
3Measurement precision
If a geometrically modified picture is generated and used for inter prediction, then prediction accuracy is improved, but the encoding and decoding process becomes more complex
Solution Approach 1:
The patent applies parameter changes by modifying geometric parameters (such as scaling factors, translation offsets, or rotation angles) of the reference picture to generate a geometrically modified picture. Instead of complex image processing, the solution changes geometric transformation parameters to align the reference picture with the current picture. This approach improves prediction accuracy while keeping the encoding/decoding complexity manageable by focusing on parameter adjustment rather than full image reprocessing.
Solution Approach 2:
The patent introduces a geometric modification process as an intermediary step between the reference picture and inter prediction. This intermediary generates a corrected reference picture that serves as a bridge, improving the match between reference and current pictures. The intermediary handles the geometric transformation in a standardized way that improves accuracy while controlling complexity through efficient transformation algorithms.
Data Source
AI summary
The present invention relates to a method and apparatus for video encoding/decoding by using a geometric modified picture. The encoding method according to the present invention includes: generating a geometric modified picture by geometrically modifying a reference picture; and generating a first prediction block of a current block by performing inter prediction referencing the geometric modified picture.