Adaptive Pre-Filtering for Video Encoding Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoders often produce poorly encoded video images with artifacts like blocking and ringing, especially in low bit rate applications with fast motion, due to limitations in pre-filtering techniques that can lead to loss of video resolution and instability in encoding processes.
Innovation Solution
An adaptive pre-filtering method using a linear filter module that extracts tuning parameters such as motion, quantization, variance, and mean square error prediction to generate control parameters, which are then used to select filter coefficients for filtering input video data before encoding, reducing encoding complexity and artifacts like blocking and ringing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If pre-filtering is applied to reduce encoding complexity and improve video quality, then encoding efficiency is improved, but video resolution is lost and image sharpness is reduced
Solution Approach 1:
The patent applies different filter strengths to different spatial regions and temporal segments of the video signal. Motion-adaptive filtering adjusts the filter coefficients based on local motion characteristics, applying stronger filtering to static regions and weaker filtering to regions with high motion or detail, thereby preserving local quality while achieving overall encoding efficiency improvement
Solution Approach 2:
The filtering parameters are made dynamic and adaptive rather than fixed. The filter strength and characteristics change dynamically based on motion estimation results, scene complexity, and encoding conditions, allowing the system to optimize between resolution preservation and encoding efficiency for each specific video segment
2Manufacturing precision
If strong pre-filtering is applied to reduce artifacts, then encoding quality is improved, but encoding stability deteriorates due to oscillations in filter response
Solution Approach 1:
The patent implements feedback mechanisms where encoding results and quality metrics are used to adjust filter parameters for subsequent processing. This closed-loop control prevents oscillations by continuously monitoring encoding quality and stabilizing filter response based on actual performance rather than theoretical optimal values
Solution Approach 2:
The system prepares multiple filter coefficient sets with varying strengths beforehand and selects appropriate pre-computed coefficients based on predicted encoding conditions. This cushioning approach prevents extreme filter responses that could cause instability by having predetermined safe filter options ready for selection
3Manufacturing precision
If motion compensated filtering is used to reduce artifacts, then video quality is improved, but system complexity increases due to additional motion estimators
Solution Approach 1:
The patent makes the existing motion estimation module perform dual functions: both standard motion compensation for prediction and motion-based filtering control. By reusing the same motion vectors and estimation results for both prediction error calculation and filter parameter selection, the system achieves motion-adaptive filtering without requiring separate motion estimation processes
4Quantity of substance
If pre-filtering is applied to reduce bit rate, then transmission efficiency is improved, but video artifacts increase due to loss of high frequency components
Solution Approach 1:
The patent dynamically changes filter parameters including cutoff frequency, filter order, and coefficient values based on scene characteristics, motion intensity, and desired bit rate targets. This parameter adaptation allows the system to achieve bit rate reduction while minimizing artifact generation by adjusting filtering aggressiveness to match content requirements
Data Source
AI summary
A method, system and apparatus for adaptive pre-filtering includes: extracting tuning parameters from video data in a video encoding process; processing the tuning parameters to generate control parameters representative of properties of the video data; coupling the control parameters to a filter response; generating filter coefficients according to the filter response; and filtering the video data using the filter coefficients prior to video encoding. The filtering of video data may thus be adapted to one or more tuning parameters of the video data and the encoding process. The filtering may decrease the complexity of encoding by an attenuation of high frequency signals, effectively decreasing the quantization step and reducing compression artifacts.