Adaptive Motion Compensation Filters for Fractional-Pixel Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing moving picture encoding and decoding methods are inefficient due to the use of fixed filters for motion compensation, which do not adapt to the bit rate and nature of the moving picture, limiting the accuracy of decimal point picture element interpolation.
Innovation Solution
The method involves detecting and switching between multiple sets of filters with different characteristics to achieve the highest encoding efficiency, using motion vectors and filter combinations to generate prediction values for improved motion compensation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If fixed filters are used for motion compensation, then device complexity is reduced, but encoding efficiency and interpolation accuracy deteriorate
Solution Approach 1:
The patent implements dynamic filter selection by switching between multiple filter types (e.g., half-sample filters and quarter-sample filters) based on the characteristics of the current picture block. The motion compensation apparatus dynamically determines which filter to use for each block, allowing the system to adapt to different picture content and motion patterns, thereby improving encoding efficiency without requiring a permanently complex filter structure for all cases.
Solution Approach 2:
The patent changes filter parameters (filter type, filter coefficients) based on picture characteristics such as motion magnitude, picture type (I-picture, P-picture, B-picture), and block size. By adjusting filter parameters according to the specific encoding context, the system achieves better interpolation accuracy and encoding efficiency while maintaining manageable device complexity through conditional parameter selection rather than permanent structural complexity.
2Measurement precision
If multiple filter sets are used for adaptive motion compensation, then interpolation accuracy is improved, but device complexity increases
Solution Approach 1:
The patent applies different filter types to different local regions (picture blocks) based on their specific characteristics. Each picture block can receive customized filter treatment according to its motion properties, picture type, and size. This local adaptation improves interpolation accuracy for each region without requiring the entire system to be maximally complex, as only the necessary filters are activated for each local area.
Solution Approach 2:
The patent divides the picture into multiple blocks and applies different filtering strategies to different blocks. By segmenting the processing into block-level decisions, the system can use simple filters for some blocks and more sophisticated filters for others, achieving high overall accuracy while keeping the average device complexity manageable through selective application of complex filtering mechanisms.
3Productivity
If filter coefficients are adaptively switched, then encoding efficiency is improved, but computational complexity increases
Solution Approach 1:
The patent pre-defines multiple filter coefficient sets corresponding to different filter types and picture conditions. These filter coefficients are prepared in advance and stored in the system, so that during encoding, the system only needs to select from the pre-prepared coefficients rather than calculating them in real-time. This preliminary preparation reduces the computational complexity during the actual encoding process while still enabling adaptive filter switching for improved encoding efficiency.
Data Source
AI summary
The present invention seeks to provide a solution to the problem that motion compensation in an encoding method of the prior art involves decimal point picture element interpolation by means of fixed filters and therefore cannot realize interpolation of decimal point picture elements that accords with the nature and bit rate of a moving picture. A filter/interpolated image memory (101) interpolates picture elements of decimal point positions in a reference image by means of a plurality of sets of filters, a motion vector/filter detector (102) detects the combination of filter and motion vector that realizes the highest encoding efficiency, a prediction value generator (103) uses the detected combination of filter and motion vector to generate a prediction value, and motion vector/filter detector (102) supplies as output the information of the filter and information of the motion vector that generated the prediction value; whereby decimal point picture element interpolation that corresponds to the nature and bit rate of the moving picture can be realized and encoding of a moving picture can be improved.