Video Quality Control via DISTS-Quantization Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies, such as VVC, face challenges in achieving optimal Rate-Distortion (R-D) performance and perceptual quality due to high computational complexity and quality fluctuation, especially in high-resolution and fast-frame-rate videos.
Innovation Solution
A Deep Image Structure and Texture Similarity (DISTS) based quality control scheme is developed, which establishes a DISTS-Quantization (D-Q) model to determine target frame-level quality and allocates coding parameters adaptively, ensuring optimal R-D performance and perceptual quality by integrating the D-Q model with the Versatile Video Coding (VVC) encoder.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If advanced coding tools are adopted to improve compression efficiency, then video quality is improved, but computational complexity increases significantly
Solution Approach 1:
The patent establishes a DISTS-quantization model in advance that predicts perceptual quality metrics based on quantization parameters. This preliminary modeling allows the encoder to estimate quality outcomes without performing full encoding, enabling quality-based decision making before actual compression operations are executed.
Solution Approach 2:
The patent introduces a quality control module that acts as an intermediary between the quantization parameter selection and the actual encoding process. This module uses the pre-established DISTS model to mediate quality requirements and automatically adjust coding parameters, reducing the need for complex trial-and-error encoding while maintaining perceptual quality.
2Manufacturing precision
If quality control is strengthened to ensure optimal R-D performance, then perceptual quality is improved, but encoding time increases
Solution Approach 1:
The patent implements a self-service quality control mechanism where the encoder automatically adjusts its own coding parameters based on the DISTS model predictions. The system monitors its own encoding state and autonomously optimizes quantization parameters to meet target quality levels without external intervention or complex external quality assessment systems.
Solution Approach 2:
The patent dynamically changes coding parameters (particularly quantization parameters) based on predicted DISTS quality metrics. By continuously adjusting these parameters according to the established model and actual encoding feedback, the system maintains optimal rate-distortion performance while adapting to different content characteristics and quality requirements.
3Stability of the object's composition
If frame-level quality control is implemented, then quality consistency is improved, but processing complexity increases
Solution Approach 1:
The patent segments the video encoding process into frame-level units, applying quality control independently to each frame based on its specific characteristics. This segmentation allows the system to maintain quality consistency across the entire video sequence while adapting to frame-specific content variations, rather than applying a single global quality setting.
Solution Approach 2:
The patent implements dynamic quality control where the target DISTS quality value and coding parameters are adjusted on a frame-by-frame basis according to content complexity and importance. This dynamic approach maintains quality consistency through adaptive adjustment rather than rigid fixed settings, allowing the system to respond to changing video content while maintaining overall quality standards.
Data Source
AI summary
There is provided a computer-implemented method for processing a video. The computer-implemented method includes: (a) determining a target frame-level quality required for a frame of the video to be encoded, the determining of the target frame-level quality is based on, at least, a rate-quantization (R-Q) model that relates bit-rate and quantization step size and a quality-quantization model that relates quality measure and the quantization step size; and (b) determining one or more coding parameters for encoding the frame based on the determined target frame-level quality.