Video Encoder Rate Control Using Noise-Aware Quantization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern video encoders like VVenC exhibit suboptimal performance on camera-captured natural content due to misclassification of frames, lack of noise-dependent rate control, and inefficient recovery from noise-induced rate savings, leading to reduced PSNR and encoding inefficiency.
Innovation Solution
Implementing a reactivation mechanism for motion compensated temporal pre-filtering on noisy frames, integrating noise-level dependent rate control, and adjusting quantization parameters to optimize encoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If MCTPF is deactivated on frames classified as SCC, then encoding speed is improved, but coding performance deteriorates on noisy natural content frames
Solution Approach 1:
The patent changes the parameter used for frame classification from a simple SCC detector output to a noise level parameter derived from motion estimation data. This allows the system to adaptively adjust the central value (threshold) for MCTPF activation based on the actual noise characteristics of each frame, thereby improving coding performance on noisy natural content while maintaining encoding speed through efficient parameter-based decision making
Solution Approach 2:
The patent performs preliminary noise level assessment using motion estimation data before making the MCTPF activation decision. By calculating the central value from motion estimation results in advance, the system prepares the necessary parameter information to quickly determine whether MCTPF should be applied, avoiding the need for complex post-processing analysis and maintaining encoding speed
2Device complexity
If RC uses fixed QPstart and chigh values, then rate control implementation is simplified, but encoding efficiency deteriorates due to instability on noisy content
Solution Approach 1:
The patent introduces dynamic adaptation into the rate control system by allowing QPstart and chigh values to be adjusted based on the noise level parameter. Instead of using fixed values throughout encoding, the system dynamically modifies these parameters according to the actual noise characteristics detected in the input content, thereby stabilizing rate control on noisy content while maintaining reasonable implementation complexity through parameter-based adaptation
Solution Approach 2:
The patent implements feedback mechanisms where the noise level parameter derived from motion estimation data is used to adjust rate control parameters. The system continuously monitors the input content characteristics and feeds this information back to modify QPstart and chigh values, creating a closed-loop control system that adapts to varying noise conditions and improves encoding efficiency
3Loss of substance
If noise-aware RC limits QP to avoid noise components, then bit budget is saved on noise, but target rate cannot be reached due to inefficient bit reallocation
Solution Approach 1:
The patent applies local quality adjustment by modifying quantization parameters selectively based on the noise level parameter. Instead of uniformly limiting QP across all blocks, the system adjusts QP values locally according to the noise characteristics of different regions and blocks, allowing efficient bit reallocation from noisy areas to important content areas, thereby achieving both noise suppression and target rate compliance
Data Source
AI summary
An apparatus and method determine quantization parameter for a second rate-control pass of a video encoder for a group of pictures based on an encoding bitrate and further quantization parameter of a first rate-control pass of the video encoder for the group of pictures. The apparatus is configured to determine, for each of positions spatially distributed over frames of the group of pictures, a dissimilarity measure or a similarity measure of a comparison of a first frame portion at the respective position, and a second portion of a reference frame to which a motion vector of an estimated motion field at the respective position points. The apparatus is configured to form a representative scalar value based on the dissimilarity measures or similarity measures determined for the positions. The apparatus is further configured to determine a preliminary quantization parameter for the second rate-control pass based on the encoding bitrate.