Region-Based In-Loop Filtering for Lower-Latency Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding and decoding technologies face inefficiencies in reducing coding artifacts through in-loop filtering, particularly due to the need for parsing filter parameters at the beginning of each slice or picture, which introduces latency and suboptimal use of filter parameters across different regions.
Innovation Solution
Implementing region-based in-loop filtering by defining filter regions within slices or pictures, allowing filter blocks to inherit parameters from neighboring blocks and sharing parameters across multiple filters, reducing the need for repetitive parameter parsing and enabling parallel processing of filtering stages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If filter parameters are parsed at the beginning of each slice or picture, then filtering can be performed on reconstructed blocks, but latency increases and computational overhead increases
Solution Approach 1:
The patent applies preliminary action by determining and storing filter parameters for multiple regions in advance before actual filtering is needed. The encoder divides the picture into multiple regions and determines filter parameters for each region ahead of time, allowing the decoder to retrieve and apply these pre-determined parameters without waiting to parse them during the filtering process, thus reducing latency.
Solution Approach 2:
The patent applies segmentation by dividing the picture into multiple distinct regions, where each region can have its own independently determined filter parameters. This allows different parts of the picture to be processed with region-specific optimization, and the parameters for each segment can be prepared and stored separately, enabling parallel processing and reducing overall latency.
2Reliability
If filter parameters are parsed at the beginning of each slice or picture, then filtering can be performed, but computational overhead increases
Solution Approach 1:
The encoder performs the computationally intensive filter parameter determination in advance and stores these parameters for later use. This shifts the computational burden from the decoder to the encoder, allowing the decoder to simply retrieve and apply pre-computed parameters, significantly reducing the computational overhead during decoding and filtering operations.
Solution Approach 2:
The patent employs copying by storing determined filter parameters in a buffer or memory structure that can be efficiently retrieved and reused. Instead of re-computing parameters during filtering, the system copies pre-determined parameters from storage to the filtering process, reducing computational overhead while maintaining filtering effectiveness.
3Productivity
If common filter parameters are used across multiple blocks, then parameter parsing is reduced, but adaptability to different regions decreases
Solution Approach 1:
The patent segments the picture into multiple regions, each capable of having its own filter parameters. This segmentation allows the system to balance between using common parameters (for processing efficiency) and region-specific parameters (for adaptability), as each region can be independently configured based on its characteristics while still benefiting from the overall regional structure.
Solution Approach 2:
The patent applies local quality by allowing different filter parameters to be assigned to different regions based on their specific characteristics. Each region can have optimized parameters tailored to its local content properties, while the systematic organization of these regional parameters maintains processing efficiency through structured storage and retrieval mechanisms.
Data Source
AI summary
Methods and apparatus for performing in-loop filtering in an encoder or a decoder provide regions which use a common set of filter parameters An index can be sent from an encoder to a decoder indicating which set of filter parameters is to be used for a particular region. The in-loop filter can be Sample Adaptive Offset, Adaptive Loop Filter, or any other such filter. An encoder classifies regions of a picture according to blocks using a common set of filter parameter. The classification can be in the form of a map. Filtering blocks use the common set of filter parameters for a region. A decoder parses a bitstream for a set of filter parameters and an index indicative of the filter parameter set for a region being decoded.