Video NNPF Bit-Mask Signaling for Stable Filter-Purpose Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in efficiently signaling the purpose of neural-network post-processing filters (NNPFs) due to an ad-hoc enumeration of values, leading to potential instability and logical issues.
Innovation Solution
Implement a systematic scheme for signaling NNPF purposes using bit masks and format changes, such as color format, resolution, frame rate, and bit-depth changes, to determine the NNPF's purpose, and conditionally signal format parameters based on bit masking operators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If ad-hoc enumeration of values is used to signal NNPF purposes, then the signaling method is simple to implement, but it leads to potential instability and logical issues
Solution Approach 1:
The patent changes the parameter representation method from ad-hoc enumeration to a systematic bit mask approach. Instead of using arbitrary enumerated values for NNPF purposes, the invention uses bit mask parameters where each bit represents a specific purpose or combination of purposes. This systematic parameter change eliminates logical issues while maintaining implementation feasibility through structured bit manipulation operations.
2Reliability
If a systematic scheme with bit masks is used to signal NNPF purposes, then reliability and extensibility are improved, but device complexity increases
Solution Approach 1:
The bit mask signaling mechanism serves multiple functions simultaneously: it reliably indicates NNPF purposes, allows for future extensions without breaking existing systems, and provides a unified framework for various filtering purposes. Each bit in the mask can represent a different purpose or combination of purposes, making the signaling mechanism universally applicable across different scenarios while maintaining simplicity through standard bit manipulation operations.
Data Source
AI summary
Embodiments of the present disclosure provide a solution for video processing. A method for video processing is proposed. The method comprises: performing a conversion between a video and a bitstream of the video, wherein a neural-network post-processing filter (NNPF) is applied on at least one picture associated with the video, the bitstream comprises a first indication indicating a purpose of the NNPF, one of a plurality of candidates for the purpose corresponds to a bit mask, and the bit mask is used for determining the purpose of the NNPF based on the first indication.