Neural-Network Post-Filter Signaling for Efficient Image Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution and high-quality images leads to a significant increase in transmission and storage costs due to the rise in the amount of transmitted information or bits, necessitating high-efficient image compression technology.
Innovation Solution
An image encoding/decoding method that includes determining the activation of neural-network post-filters using supplemental enhancement information messages, ensuring clear identification and efficient use of neural-network post-filters, and encoding this information into bitstreams for improved encoding/decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If high-resolution and high-quality images are transmitted, then image quality is improved, but transmission cost and storage cost increase
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the activation of neural-network post-filters based on image characteristics and coding conditions. The system changes filter activation parameters (activating or deactivating specific filters) to optimize the balance between image quality and data量, thereby reducing transmission and storage costs while maintaining acceptable quality levels.
2Productivity
If neural-network post-filters are used to improve image quality, then encoding efficiency is improved, but complexity of determining which filter to activate increases
Solution Approach 1:
The patent implements preliminary action by pre-defining multiple neural-network post-filters with different characteristics and preparing their activation conditions in advance. The system establishes a predetermined set of filters and their selection criteria before processing, which simplifies the real-time decision-making process while maintaining high encoding efficiency.
Solution Approach 2:
The patent applies dynamics by making the filter selection process adaptive and flexible. The system dynamically determines which neural-network post-filter to activate based on real-time analysis of image characteristics and coding conditions, allowing the encoding efficiency to be optimized for each specific situation while managing complexity through structured decision frameworks.
3Adaptability or versatility
If multiple neural-network post-filters are defined, then image quality improvement options increase, but ambiguity arises in determining which filter to activate
Solution Approach 1:
The patent implements feedback mechanisms that continuously monitor image characteristics and coding conditions to determine the appropriate filter activation. The system uses feedback from image analysis to resolve ambiguities in filter selection, ensuring clear and unambiguous determination of which filter to activate while maintaining high adaptability and versatility.
Data Source
AI summary
An image decoding method includes: acquiring a neural-network post-filter characteristics (NNPFC) SEI message and a neural-network post-filter activation (NNPFA) SEI message; based on the NNPFC SEI message, determining at least one neural network which may be used as a neural-network post-processing filter; and based on the NNPFA SEI message, determining whether to activate a target neural-network post-processing filter that may be applied to the current picture. The NNPFA SEI message includes target identification information and target basic flag information of the target neural-network post-processing filter, the target neural-network post-processing filter is determined based on the target identification information and the target basic flag information, and based on that the target basic flag information does not indicate the target neural-network post-processing filter is a basic neural-network post-processing filter, at least one NNPFC SEI message is present before the NNPFA SEI in decoding order.