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 a neural-network post-filter (NNPF) based on supplemental enhancement information (SEI) messages, specifying which neural-network post-processing filter to apply, and encoding this information in bitstreams for improved encoding/decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If high-resolution and high-quality image data is transmitted, then image quality is improved, but transmission cost and storage cost increase
Solution Approach 1:
The patent extracts and transmits only the essential image data and necessary metadata (such as neural-network post-filter characteristics and activation information) while removing redundant information. This allows high-quality image reconstruction with reduced transmission volume, directly addressing the contradiction between image quality and transmission amount.
Solution Approach 2:
The patent performs preliminary compression and optimization of image data before transmission by applying neural-network post-filters during the encoding phase. This preliminary processing reduces the amount of data that needs to be transmitted while preserving image quality, resolving the contradiction between high image quality and low transmission cost.
2Productivity
If neural-network post-filter information is encoded in SEI messages, then encoding efficiency is improved, but bitstream complexity increases
Solution Approach 1:
The patent uses SEI (Supplemental Enhancement Information) messages as an intermediary layer to carry neural-network post-filter characteristics and activation information. This intermediary structure allows the transmission of complex filter information without directly increasing the core bitstream structure, thus improving encoding efficiency while managing complexity through a standardized interface.
Solution Approach 2:
The patent encodes neural-network post-filter information using compact parameter representations, such as filter identification indices and activation flags, rather than full filter definitions. This parameterization approach enables efficient encoding of filter information while maintaining manageable bitstream complexity through concise data structures.
Data Source
AI summary
An image encoding/decoding method, a bitstream transmission method, and a computer-readable recording medium on which a bitstream is stored are provided. The image decoding method according to the present disclosure comprises the steps of: acquiring a neural-network post-filter (NNPF) supplemental enhancement information (SEI) message; determining, on the basis of at least one neural-network post-filter characteristics (NNPFC) SEI message included in the NNPF SEI message, at least one neural network that can be used as a post-processing filter, on the basis that the NNPF SEI message is applied to a current picture; and determining, on the basis of at least one neural-network post-filter activation (NNPFA) included in the NNPF SEI message, whether a target neural-network post-processing filter applicable to the current picture is activated, wherein the target neural-network post-processing filter is determined as a neural-network post-processing filter of the last NNPFC SEI message, from among NNPFC SEI messages, of a decoding order.