NN Post-Filter Activation Signaling for Video Quality and Bandwidth

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video coding standards, such as VVC, lack efficient mechanisms for signaling and utilizing neural-network post-processing filters (NNPFs) to enhance video quality, leading to suboptimal video processing and increased bandwidth demand.

Innovation Solution

Implementing a method for passing activation parameters to a neural-network post-filter (NNPF) during video processing, enabling conversion between visual media data and a bitstream, and generating a bitstream based on these parameters, using SEI messages for signaling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If neural-network post-processing filters (NNPFs) are implemented to enhance video quality, then video quality is improved, but device complexity and processing overhead increase

Engineering Contradiction:

Improvevideo qualityVSAvoidprocessing complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-defining NNPF activation conditions and parameters during video encoding, storing them in the bitstream. This allows the decoder to directly apply pre-configured filtering without real-time complex decision-making, reducing processing complexity while maintaining enhanced video quality through neural network post-processing

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements parameter changes by dynamically adjusting NNPF activation parameters based on video content characteristics, such as scene complexity, motion intensity, and quality requirements. This enables selective application of neural network filtering only when beneficial, improving video quality while avoiding unnecessary processing complexity in simple scenes

Inventive Principle:

Principle #35Parameter changes

2Productivity

If NNPF activation parameters are signaled in the bitstream, then video processing efficiency is improved, but bandwidth consumption increases

Engineering Contradiction:

Improvevideo processing efficiencyVSAvoidbandwidth

Core Design Contradiction:

ProductivityVSQuantity of substance

Solution Approach 1:

The patent applies local quality by signaling NNPF activation parameters selectively for different video regions, temporal layers, or quality levels rather than uniformly for the entire video stream. This allows efficient video processing in critical regions while reducing bandwidth consumption in less important areas through parameter omission or compression

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent implements partial action by signaling only the essential NNPF parameters needed for effective filtering, omitting redundant information. This provides sufficient processing efficiency for quality enhancement while minimizing bandwidth consumption by transmitting only the necessary subset of parameters

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS20250392762A1Neural-network post-processing filter activation parameters

Publication Date: 2025.12.25 DOUYIN VISION CO LTD

AI summary

A mechanism for processing video data is disclosed. The mechanism includes determining to pass parameters to a neural-network post-filter (NNPF) when the NNPF is activated. A conversion is performed between a visual media data and a bitstream based on the parameters. The parameters may be referred to as activation parameters when the parameters are passed to the NNPF upon activation.