Perceptual Video Encoding via Macro-Block Activity Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques fail to effectively exploit the perceptual properties of the human visual system (HVS) to optimize quantization in different regions of a video frame, leading to inefficient compression and noticeable artifacts in less textured areas.
Innovation Solution
A video encoder determines an activity measure for macro-blocks by computing statistical measures and modulating the quantization scale based on these measures, classifying macro-blocks into types like smooth, edge, and texture, and applying specific quantization scales and encoding modes to adapt to the sensitivity of the HVS.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a uniform quantization scale is applied to all macro-blocks, then the encoding process is simple and fast, but visual quality deteriorates in less textured regions due to noticeable artifacts
Solution Approach 1:
The patent applies different quantization scales to different macro-blocks based on their texture characteristics. Smooth macro-blocks receive finer quantization (lower scale) while textured macro-blocks receive coarser quantization (higher scale), optimizing visual quality locally without compromising overall encoding efficiency
Solution Approach 2:
The quantization scale is dynamically adjusted for each macro-block based on its activity measure and texture content. The encoder computes statistical measures (variance, standard deviation) for each macro-block and adapts the quantization scale accordingly, rather than using a static uniform scale
2Quantity of substance
If coarse quantization is applied to all regions, then bit rate is reduced, but visual quality deteriorates in smooth regions due to noticeable artifacts
Solution Approach 1:
The patent implements region-adaptive quantization where smooth macro-blocks are identified through statistical analysis (low variance/standard deviation) and assigned lower quantization scales to preserve visual quality, while textured regions can tolerate higher quantization scales
Solution Approach 2:
The quantization scale parameter is changed dynamically based on the activity measure of each macro-block. The encoder computes statistical parameters (variance, standard deviation, zero-crossing rate) and adjusts the quantization scale to optimize the trade-off between bit rate and visual quality for each region
3Object-affected harmful factors
If fine quantization is applied to all macro-blocks, then visual quality is maintained, but bit rate increases and compression efficiency decreases
Solution Approach 1:
The patent applies fine quantization (lower scale) only to smooth macro-blocks where it is visually necessary, while textured macro-blocks use coarser quantization (higher scale) where the human visual system is less sensitive, thereby reducing overall bit rate while maintaining perceived quality
Solution Approach 2:
Instead of applying fine quantization uniformly to all macro-blocks (excessive action), the patent applies it selectively only where needed (partial action) - specifically to smooth regions with low activity measures, avoiding unnecessary bit consumption in textured regions
4Object-affected harmful factors
If region-adaptive quantization is implemented, then visual quality and compression efficiency are optimized, but device complexity increases due to multiple statistical computations
Solution Approach 1:
The patent divides the video frame into macro-blocks and further into quadrants, computing statistical measures for each region. This segmentation allows parallel computation and efficient processing of large video data while maintaining adaptive quantization benefits
Solution Approach 2:
The patent replaces complex perceptual quality assessment with simpler statistical computations (variance, standard deviation, zero-crossing rate) that correlate with texture content. This substitution maintains adaptive quantization effectiveness while significantly reducing computational complexity
Data Source
AI summary
In a video encoder, pixel values of a macro-block are processed to determine an activity measure indicative of the type of content in the macro-block. Several techniques are employed for determining the activity measure of a macro-block. In an embodiment, a default quantization scale for quantizing a macro-block is modified based on the activity measure of the macro-block. In another embodiment, the macro-block is classified into one of multiple classes based on its activity measure. The default quantization scale for quantizing the macro-block is modified based on the classification of the macro-block. In yet another embodiment, an encoding mode to be used for encoding a macro-block is also determined on the basis of the class of the macro-block. Several of the techniques exploit the fact that the human visual system (HVS) has different sensitivities in perceiving a (rendered) macro-block or video frame, depending on the type of macro-block content.