Scalable Nesting SEI Message for Video Picture Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current scalable video coding standards, such as H.264/AVC, lack a mechanism to effectively associate supplemental enhancement information (SEI) messages with specific pictures within an access unit, limiting their applicability and scope, especially in scalable video, multiview video, and multiple description coding scenarios.
Innovation Solution
The introduction of a scalable nesting SEI message that indicates which pictures within an access unit a supplemental enhancement information message applies to, allowing for the reuse of H.264/AVC SEI messages with minor semantic additions, enabling their application across scalability layers, views, and multiple description coding descriptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If H.264/AVC SEI messages are used without picture association mechanism, then the decoding process can generate correct sample values, but the applicability and scope of SEI messages is limited in scalable video and multiview video scenarios
Solution Approach 1:
The patent segments the SEI message handling by introducing picture association indicators that divide and specify which particular pictures within an access unit each SEI message applies to. This segmentation allows selective application of SEI messages to specific pictures (e.g., base layer vs. enhancement layer pictures) rather than requiring universal handling across all pictures, thereby improving adaptability while maintaining manageable complexity through structured organization.
Solution Approach 2:
The patent applies local quality by allowing different SEI message associations for different pictures within the same access unit. Each picture can have specific SEI messages associated with it based on layer type, view index, or other identifying characteristics. This enables tailored enhancement information delivery to specific pictures (e.g., applying error resilience SEI messages only to base layer pictures or specific enhancement layer pictures) rather than uniformly to all pictures, improving versatility without proportionally increasing complexity.
2Adaptability or versatility
If SEI messages are extended to apply to all pictures in an access unit, then scalability and flexibility are enhanced, but the complexity of message association and processing increases
Solution Approach 1:
The patent implements preliminary action by pre-defining picture identification mechanisms (such as layer type indicators, view indices, and picture identification fields) that are established before SEI message processing occurs. These preliminary structures enable efficient association of SEI messages with appropriate pictures during decoding without requiring complex real-time analysis, thus achieving enhanced scalability while controlling processing complexity through advance organization of picture metadata.
3Measurement precision
If picture association indicators are added to SEI messages, then precise association with encoded pictures is achieved, but the data structure complexity increases
Solution Approach 1:
The patent extracts picture identification information from the existing scalable video coding structure (such as layer type, view index, temporal ID, and picture order count) and uses these extracted fields as association indicators for SEI messages. Rather than creating entirely new complex data structures, the patent leverages already-present picture identification mechanisms, thereby achieving precise SEI message association with encoded pictures while minimizing additional data structure complexity by reusing existing standardized fields.
Data Source
AI summary
A system and method for conveying supplemental enhancement information related to different pictures in one access unit. Different pictures in one access unit may be pictures of each layer of a scalable video, pictures of each view of a multiview video, or pictures of each description in a multiple description coding (MDC) video. An indication is included in the coded bitstream indicating the pictures with which the supplemental enhancement information message is associated. During a subsequent decoding process, the device performing the decoding recognizes this information and uses it appropriately.