Joint PSAO Parameter Selection for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression systems, such as HEVC and VVC, face inefficiencies in selecting Pre-Sample Adaptive Offset (PSAO) parameters for in-loop filtering, which can lead to suboptimal compression performance due to sequential and local optimization processes that do not account for the joint impact of deblocking and PSAO filtering.
Innovation Solution
A method for jointly selecting PSAO parameters for vertical and horizontal filtering by minimizing a delta cost that combines delta distortion and delta rate, ensuring more precise adaptation and overcoming the limitations of separate optimization for PSAOV and PSAOH filters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If sequential and local optimization processes are used for selecting PSAO parameters, then the parameter selection process is simpler, but compression performance becomes suboptimal
Solution Approach 1:
The patent combines the separate optimization processes for PSAOV and PSAOH parameters into a single joint optimization process. The encoder simultaneously selects both parameter sets by minimizing a combined cost function that includes distortion and rate components for both vertical and horizontal filtering, rather than optimizing them sequentially and independently.
Solution Approach 2:
The patent changes the optimization approach from local sequential parameter selection to global joint parameter selection. The encoder evaluates multiple candidate combinations of PSAOV and PSAOH parameters together and selects the combination that minimizes the overall delta cost, allowing the system to find globally optimal parameter settings rather than getting stuck in local optima.
2Device complexity
If separate optimization is performed for PSAOV and PSAOH filters, then the computational complexity is reduced, but the joint impact of deblocking and PSAO filtering is not accounted for
Solution Approach 1:
The patent merges the separate optimization processes for PSAOV and PSAOH filters into a unified joint optimization process. The encoder simultaneously considers both parameter sets and selects them together based on a combined cost function, ensuring that the joint impact of deblocking and PSAO filtering is properly accounted for in the optimization.
Solution Approach 2:
The patent implements a feedback mechanism where the encoder evaluates the combined effect of deblocking filtering and PSAO filtering on the reconstructed picture quality. By calculating the delta cost that incorporates both vertical and horizontal filtering impacts, the system uses this feedback to guide the joint selection of PSAO parameters, ensuring that the optimization accounts for the interaction between different filtering stages.
3Reliability
If joint selection of PSAO parameters is performed, then compression performance is improved, but the complexity of parameter selection increases
Solution Approach 1:
The patent segments the joint optimization process into manageable components by defining a structured cost function that separately calculates delta distortion and delta rate for both PSAOV and PSAOH parameters. The encoder evaluates candidate parameter combinations using this segmented approach, computing the impact of each parameter set on picture quality and bitrate independently before combining them into an overall cost metric for comparison.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The present application relates to a method of encoding an original video picture comprising pre-Sample Adaptive Offset vertical filtering using a first combination of PSAO parameters and pre-Sample Adaptive Offset horizontal filtering using a second combination of PSAO parameters intended to be applied on video picture data, wherein the first and second combinations of PSAO parameters are selected jointly by minimizing a delta cost (ΔJi,j) defined as a linear combination of a delta distortion and a delta rate calculated for a couple of a first candidate combination of PSAO parameters and a second candidate combination of PSAO parameters, said delta distortion equals to a difference between a first distortion calculated by taking into account the pre-Sample Adaptive Offset vertical and horizontal filtering and a second distortion calculated without applying the pre-Sample Adaptive Offset vertical and horizontal filtering.