Block-Size Adaptive Transform Basis Selection in Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding methods require excessive computational load and time due to the evaluation of multiple transform bases for encoding and decoding, especially for varying block sizes, leading to increased circuit scale and processing time.
Innovation Solution
An encoder and decoder that adaptively select transform and inverse transform basis candidates based on block size, reducing the number of candidates and excluding those requiring high computation, thereby optimizing processing load and time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple transform basis candidates are evaluated for all block sizes, then transform accuracy is improved, but computational load and processing time increase
Solution Approach 1:
The patent applies parameter changes by adapting the number of transform basis candidates based on block size parameters. For small blocks (4×4, 8×8), only one transform basis candidate is used, while for large blocks (16×16, 32×32), multiple candidates are evaluated. This parameter-based adaptation resolves the contradiction by optimizing the balance between transform accuracy and processing speed according to different block size conditions.
2Measurement precision
If multiple transform basis candidates are evaluated for all block sizes, then transform accuracy is improved, but circuit scale increases
Solution Approach 1:
The patent implements local quality by applying different transform basis candidate strategies to different block size regions. Small blocks use a simplified single-candidate approach, while large blocks use the multi-candidate evaluation approach. This localized differentiation reduces overall circuit complexity while maintaining transform accuracy where it is most needed (in large blocks that contribute more to overall image quality).
3Ease of operation
If transform basis candidates are selected without considering block size, then processing is simplified, but encoding efficiency decreases
Solution Approach 1:
The patent applies dynamics by making the transform basis candidate selection process adaptive rather than static. The number of candidates is dynamically adjusted based on block size: one candidate for small blocks and multiple candidates for large blocks. This dynamic adaptation improves encoding efficiency by optimizing the trade-off between processing simplicity and transform accuracy for different block sizes, rather than using a fixed approach for all blocks.
Data Source
AI summary
An encoder which transforms a current block to be encoded in an image to encode the current block includes circuitry and memory. The circuitry, using the memory: determines a plurality of first transform basis candidates and transforms the current block using a transform basis included in the plurality of first transform basis candidates determined, when the current block has a first size; and determines one or more second transform basis candidates different from the plurality of first transform basis candidates and transforms the current block using a transform basis included in the one or more second transform basis candidates determined, when the current block has a second size larger than the first size.