SEI-Guided Neural Post-Filter Selection in 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, 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 the bitstream to improve 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 increases
Solution Approach 1:
The patent applies preliminary action by performing neural-network post-filtering on the encoded bitstream before transmission. This pre-processing step enhances image quality in advance, allowing for more efficient compression and reduced transmission requirements without sacrificing quality.
Solution Approach 2:
The patent replaces traditional mechanical filtering methods with neural-network-based post-filtering. This substitution enables more effective image quality enhancement with lower computational overhead and reduced transmission requirements, resolving the contradiction between quality and transmission cost.
2Adaptability or versatility
If multiple NNPFC SEI messages are used to specify different neural-network post-filters, then filter selection flexibility is improved, but decoding complexity increases
Solution Approach 1:
The patent segments the filter selection process into distinct NNPFC SEI messages, each specifying different neural-network post-filter characteristics. This segmentation allows flexible filter selection while organizing the complexity into manageable, independently decodable units.
Solution Approach 2:
The patent implements dynamic filter selection where the decoder can adaptively choose which NNPFC SEI message to apply based on current picture characteristics. This dynamic approach provides flexibility while allowing the decoder to optimize computational resources by selecting only the necessary filter specifications.
Data Source
Figure 1
Figure 2
Figure 3
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.