NNPF SEI Signaling for Video Color Space Range Conversion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding standards, such as VVC, do not adequately address the need for efficient signaling and processing of neural-network post-filter (NNPF) output pictures when color spaces between NNPF output and decoded pictures differ, leading to potential issues in video decoding and display.

Innovation Solution

A method and apparatus for determining and signaling a syntax element to indicate whether NNPF output pictures are in a full range when the color spaces of NNPF output and decoded pictures differ, and performing a conversion between visual media data and a bitstream based on this determination.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If neural-network post-filter is applied to improve video quality, then video decoding and display quality is improved, but color space discrepancies between NNPF output and decoded pictures cause processing issues

Engineering Contradiction:
Improvevideo decoding and display qualityVSAvoidcolor space compatibility
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The patent applies preliminary action by determining the full range flag for NNPF output pictures in advance, before actual video decoding and display processing. This allows the system to pre-establish the correct color space parameters and conversion requirements, ensuring that when the NNPF output is processed, the color space discrepancies are already accounted for and resolved through appropriate conversion operations.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If color space conversion is performed to resolve discrepancies, then color space compatibility is improved, but processing complexity increases

Engineering Contradiction:
Improvecolor space compatibilityVSAvoidprocessing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts the color space conversion requirement as a separate, identifiable parameter (the full range flag) from the overall video processing pipeline. By isolating this specific conversion need and signaling it through the bitstream, the system can apply targeted color space conversion only when necessary, rather than performing complex general-purpose color space transformations for all NNPF output, thus reducing overall processing complexity.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If syntax element signaling is added to indicate full range information, then color space accuracy is improved, but bitstream overhead increases

Engineering Contradiction:
Improvecolor space accuracyVSAvoidbitstream overhead
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent employs a simple, single-bit syntax element (full range flag) to convey essential color space information, rather than transmitting complete color space transformation matrices or extensive metadata. This lightweight signaling approach provides sufficient color space accuracy for proper processing while minimizing bitstream overhead, effectively using a simple, disposable indicator rather than complex, redundant data structures.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Data Source

PatentUS20250386057A1Video usability information related indications and miscellaneous items in neural-network post-processing filter SEI messages
Publication Date: 2025.12.18 BYTEDANCE INC
  • US20250386057A1 patent drawing
  • US20250386057A1 patent drawing
  • US20250386057A1 patent drawing

AI summary

A mechanism for processing video data is disclosed. The mechanism includes determining to signal a syntax element to indicate whether neural-network post-filter (NNPF) output pictures are in a full range when a color space of the NNPF output pictures and a color space of decoded pictures or a color space of cropped decoded output pictures are different. A conversion is performed between a visual media data and a bitstream based on the NNPF.