Transform Selection for Non-Baseband Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression techniques introduce artifacts when encoding non-baseband signals with sharp edges and staircase changes, such as computer-generated content, due to the use of sinusoidal-based transforms like DCT, which are not well-suited for capturing these features.
Innovation Solution
Selecting between sinusoidal-based and staircase-based transforms for encoding non-baseband video signals based on rate-distortion values to optimize compression efficiency and reduce artifacts, with the staircase-based transform being used to better capture sharp edges and staircase changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If sinusoidal-based transforms (e.g., DCT) are used for video compression, then compression efficiency is improved, but artifacts (ringing effects) are introduced when encoding non-baseband signals with sharp edges and staircase changes
Solution Approach 1:
The patent changes the transform type parameter from sinusoidal-based (DCT) to staircase-based transforms when encoding non-baseband signals. This parameter change allows the encoding process to adapt to the signal characteristics, reducing ringing artifacts while maintaining compression efficiency for signals with sharp edges and staircase changes.
Solution Approach 2:
The patent implements dynamic transform selection that adapts to the signal type being encoded. By detecting whether the signal is baseband or non-baseband and selecting the appropriate transform type accordingly, the system dynamically optimizes both compression efficiency and artifact reduction based on the specific encoding requirements.
2Device complexity
If a single transform type is used for all video signals, then device complexity is reduced, but encoding precision deteriorates for non-baseband signals with sharp edges
Solution Approach 1:
The patent introduces a transform type parameter that can be changed based on signal characteristics. By selecting between sinusoidal-based and staircase-based transforms according to the signal type, the system achieves high encoding precision for non-baseband signals without requiring overly complex device architecture.
Solution Approach 2:
The patent segments the video signal processing into different transform types based on signal characteristics. By dividing the encoding process into baseband signal encoding (using sinusoidal transforms) and non-baseband signal encoding (using staircase transforms), the system achieves high precision for each segment while keeping overall complexity manageable.
3Quantity of substance
If compression techniques are applied to reduce video data size, then transmission and storage efficiency is improved, but visual quality deteriorates due to introduced artifacts
Solution Approach 1:
The patent changes the transform parameter based on signal type to minimize artifact generation during compression. For non-baseband signals with sharp edges, staircase-based transforms are used which reduce ringing artifacts while maintaining compression efficiency, thus preserving visual quality even as data size is reduced.
Solution Approach 2:
The patent converts the potential harm of compression artifacts into a benefit by selecting transform types that are specifically suited for the signal characteristics. Staircase-based transforms, when applied to non-baseband signals, transform what would be harmful ringing artifacts into beneficial sharp edge preservation, improving visual quality while maintaining compression.
Data Source
AI summary
Blocks of a frame of a non-baseband video signal are encoded by selecting between a staircase- or sinusoidal-based transform. For each prediction mode, a prediction block for encoding a current block is generated, a residual block is generated, the residual block is transformed using a sinusoidal-based transform and using a staircase-based transform, and first and second rate-distortion value for encoding the residual block using the sinusoidal-based transform and staircase-based transform are respectively generated. One of the staircase-based transform or the sinusoidal-based transform for encoding the residual block is selected based on a lowest value of the first rate-distortion value and the second rate-distortion value among results for each prediction mode, and the current block is encoded using the selected one of the staircase-based transform or the sinusoidal-based transform. Improved appearance for blocks having sharp edges can be achieved and lossless encoding is possible using the staircase-based transform.