Transform Normalization Signaling for Neural Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face challenges in efficiently encoding and decoding video data, particularly in handling intra and inter prediction techniques, transform techniques, and entropy coding, which can lead to suboptimal compression and decoding performance.
Innovation Solution
Implementing a logit transform normalization technique and sigmoid transform normalization technique in video coding, using a slope adjustment coefficient to optimize the encoding and decoding process, and generating feature data for input into a neural network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional video coding standards (H.264, H.265, H.266) are used for encoding and decoding video data, then compatibility and widespread support are maintained, but compression efficiency and decoding performance are suboptimal
Solution Approach 1:
The patent applies parameter changes by modifying the normalization technique parameters in video coding. Specifically, it introduces a logit transform normalization technique with adjustable slope adjustment coefficients, transforming the traditional normalization approach to achieve better compression efficiency while maintaining compatibility with existing video coding standards through controlled parameter modification.
Solution Approach 2:
The patent implements dynamics by making the normalization technique adaptive and configurable. The slope adjustment coefficient allows the system to dynamically adjust the normalization behavior based on content characteristics, enabling optimal compression performance while maintaining flexibility to work with different video coding standards and requirements.
2Manufacturing precision
If existing intra and inter prediction techniques are used, then standard video coding processes are maintained, but compression performance and decoding quality are suboptimal
Solution Approach 1:
The patent applies preliminary action by performing logit transform normalization on prediction residues before the main encoding process. This preprocessing step transforms the residue data to enhance the effectiveness of subsequent quantization and transform operations, thereby improving decoding quality without requiring fundamental changes to the overall coding architecture.
Solution Approach 2:
The patent introduces an intermediary normalization process between prediction and transform stages. The logit transform acts as an intermediary that processes prediction residues, adjusting their distribution characteristics to improve overall coding efficiency and decoding quality while maintaining compatibility with existing coding frameworks.
Data Source
AI summary
A device may be configured to receive a bitstream. The device may decode the bitstream according to a video coding technique. The device may determine a normalization technique from one or more normalization techniques. The device may perform the normalization technique on the decoded bitstream to generate feature data for input into a neural network. Normalizations techniques may include a combination of a linear normalization techniques, scaling down normalization techniques, and logarithmic normalization techniques. In one example, a normalization technique may include a transform based normalization technique.