Neural-Network Post-Filter Color Control Under Energy Constraints

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding systems lack efficient methods for color manipulation that balance quality and energy consumption, leading to suboptimal performance in video decoding and encoding processes.

Innovation Solution

Implementing a neural-network post-filter (NNPF) for color manipulation, which is controlled by metadata parameters and energy modification indications, allowing dynamic adjustment based on detected events and energy consumption adaptation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If neural-network post-filter is applied for color manipulation, then video quality is improved, but energy consumption increases

Engineering Contradiction:
Improvevideo qualityVSAvoidenergy consumption
Core Design Contradiction:
Manufacturing precisionVSUse of energy by moving object

Solution Approach 1:

The patent implements dynamic control of the neural-network post-filter through metadata parameters that allow the filter to be adaptively applied or adjusted based on content characteristics and energy conditions. The system can dynamically switch between applying the filter for high quality and bypassing it for energy conservation, resolving the contradiction between quality and energy consumption.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent uses metadata parameters to control the application of the neural-network post-filter, allowing dynamic adjustment of processing parameters based on energy events and content characteristics. This enables the system to change operational parameters to balance quality and energy consumption according to specific conditions.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If color manipulation is applied to all video content, then visual quality is enhanced, but processing complexity increases

Engineering Contradiction:
Improvevisual qualityVSAvoidprocessing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies color manipulation selectively to specific video content based on metadata parameters and content characteristics rather than uniformly to all content. This local application approach enhances visual quality where needed while avoiding unnecessary processing complexity for content that does not require enhancement.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent implements partial application of the neural-network post-filter based on energy events and metadata indications. Instead of applying full processing to all content, the system applies processing only when and where it provides value, reducing overall processing complexity while maintaining quality enhancement where beneficial.

Inventive Principle:
Principle #16Partial or excessive action

3Use of energy by moving object

If neural-network post-filter with energy modification is applied, then energy efficiency is improved, but video processing time increases

Engineering Contradiction:
Improveenergy efficiencyVSAvoidvideo processing time
Core Design Contradiction:
Use of energy by moving objectVSLoss of time

Solution Approach 1:

The patent implements periodic or event-driven application of the neural-network post-filter based on energy events detected during video processing. Instead of continuous processing, the system periodically evaluates energy conditions and applies the filter selectively, improving energy efficiency while minimizing additional processing time through targeted rather than continuous operation.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentEP4676061A1Neural network post filter for color manipulation
Publication Date: 2026.01.07 INTERDIGITAL CE PATENT HOLDINGS SAS
  • EP4676061A1 patent drawingFigure 1A
  • EP4676061A1 patent drawingFigure 1B
  • EP4676061A1 patent drawingFigure 1C

AI summary

A video encoding device may determine to apply neural-network post-filter (NNPF) for color manipulation. The device may, based on a determination to apply NNPF, obtain at least one metadata parameter associated with NNPF color manipulation. The device may include, in video data, a neural-network post-filter color manipulation indication to indicate that NNPF for color manipulation is to be applied. A video decoding device may determine to apply NNPF for color manipulation based on a neural-network post-filter color manipulation indication. The device may, based on a determination to apply NNPF, determine at least one metadata parameter associated with NNPF color manipulation. The device may apply NNPF for color manipulation based on the at least one metadata parameter. A device may obtain an NNPF characteristics (NNPFC) color manipulation sub-purpose indication. The at least one metadata parameter associated with NNPF color manipulation may be determined based on the NNPFC color manipulation sub-purpose indication.