Neural-Network Post-Filter SEI Signaling for Picture Rate Upsampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards lack efficient methods for signaling neural-network post-filter purposes, particularly for picture rate upsampling, which affects the processing and display of digital video data.
Innovation Solution
Implementing a neural-network post-filter (NNPF) with enhanced signaling through supplemental enhancement information (SEI) messages to manage picture rate upsampling, allowing for more efficient conversion and processing of video data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing video coding standards are used without enhanced signaling methods, then the processing of digital video data follows conventional approaches, but the efficiency of neural-network post-filter signaling is insufficient
Solution Approach 1:
The patent introduces SEI (Supplemental Enhancement Information) messages as an intermediary mechanism to carry neural-network post-filter purpose information. This allows the filtering purpose to be signaled separately from the main video bitstream, enabling efficient communication of filter characteristics without disrupting the existing video coding structure. The SEI message acts as a mediator between the encoder's filtering decisions and the decoder's reconstruction process.
2Reliability
If neural-network post-filter signaling is enhanced with SEI messages, then the processing and display of digital video data is improved, but the device complexity increases
Solution Approach 1:
The patent segments the video coding process by separating the neural-network post-filter signaling from the main video bitstream. The filtering purpose and characteristics are encoded in distinct SEI messages that are processed independently from the core video data. This segmentation allows the complexity of neural-network filtering to be managed separately, enabling reliable video processing without significantly increasing the complexity of the main coding/decoding path.
Data Source
AI summary
A mechanism for processing video data is disclosed. The mechanism includes determining a neural-network post-filter (NNPF) purpose based on a neural-network post-filter characteristics (NNPFC) supplemental enhancement information (SEI) message, wherein the NNPF purpose includes two or more types of post-filter operations. A conversion is performed between a visual media data and a bitstream based on the NNPF purpose.