Shared Reference Picture Lists for Multi-View Bitrate Savings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies struggle to efficiently compress multi-view video content, leading to high bandwidth and storage requirements due to the lack of effective methods for reducing statistical redundancy between different views.
Innovation Solution
A disparity-compensated prediction approach is implemented, where pictures from different views are included in the reference picture list simultaneously, allowing for shared reference picture information and reducing the need for independent encoding and decoding of each view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If simulcast coding is used to transmit each view independently, then decoding simplicity is maintained, but bandwidth and storage requirements increase significantly
Solution Approach 1:
The patent merges the reference picture lists of multiple views into a single shared reference picture list. Instead of maintaining separate reference picture lists for each view (as in simulcast coding), the system combines them to create a common reference structure that is accessed by multiple views, thereby reducing the total amount of data that needs to be transmitted and stored while maintaining decoding capability.
Solution Approach 2:
The shared reference picture list serves multiple views simultaneously, making the reference picture structure universal. A single reference picture list is used by multiple different views to perform inter-view prediction, allowing the same reference structure to serve multiple functions (supporting multiple views) rather than requiring separate dedicated reference lists for each view.
2Adaptability or versatility
If independent encoding is used for each view, then view-specific optimization is maintained, but statistical redundancy between views is not reduced
Solution Approach 1:
The patent combines multiple view-specific reference picture lists into a single shared reference picture list that is accessed by multiple views. This merging allows the system to exploit statistical redundancy between views by using common references, thereby reducing the total bitrate required while still allowing each view to be decoded independently using the shared references.
3Productivity
If shared reference picture information is used across views, then coding efficiency and bitrate savings are achieved, but reference picture list management complexity increases
Solution Approach 1:
The patent segments the reference picture list management into two distinct parts: a base reference picture list that is common to all views, and view-specific delta reference picture lists that contain only the additional references needed for each particular view. This segmentation reduces the overall complexity by avoiding the need to manage completely separate reference lists for each view while still allowing efficient shared access to the common reference structure.
Data Source
AI summary
The various implementations described herein include methods and systems for coding video. In one aspect, a video decoding method includes receiving a multi-view video bitstream comprising a plurality of pictures. The plurality of pictures includes a first picture corresponding to a first view and a second picture corresponding to a second view. The method includes determining whether reference picture information is shared for the first picture and the second picture. The method includes, when the reference picture information is shared, decoding the first picture and the second picture using the reference picture information.