Depth Map Encoding via Sparse Dyadic Partitioning and Adaptive Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video coding techniques result in large artifacts around sharp edges in depth maps, making it costly to faithfully represent depth edges, which is essential for rendering high-quality virtual views in 3D video applications like 3DTV and FVV.
Innovation Solution
The implementation of Sparse Dyadic Mode and joint bilateral filtering for depth map coding, which simplifies the representation of depth variations and edges, allowing for efficient encoding while maintaining rendering quality by using edge information from 2D video frames and adaptive weighting based on distance, depth difference, and image difference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional video coding techniques are used to encode depth maps, then encoding simplicity is maintained, but large artifacts appear around sharp edges and rendering quality deteriorates
Solution Approach 1:
The depth map is divided into multiple partitions, and each partition is further divided into sub-partitions. This segmentation allows different encoding strategies to be applied to different regions, particularly enabling refined partitioning around depth edges to preserve edge quality while reducing artifacts in flat regions.
Solution Approach 2:
The patent applies adaptive weighting based on local characteristics including distance, depth difference, and image difference. This allows the encoding to maintain high quality in regions with depth edges while using more aggressive compression in flat regions, thereby reducing overall artifacts while preserving critical edge information.
2Manufacturing precision
If depth edges are faithfully represented to improve rendering quality, then rendering precision is improved, but bit rate increases significantly
Solution Approach 1:
The patent employs dynamic partition refinement where the partition structure adapts to the local depth variations. In regions with sharp depth edges, partitions are refined to capture edge details, while in flat regions, coarser partitions are used. This dynamic adaptation allows faithful edge representation only where necessary, controlling bit rate while maintaining rendering quality.
Solution Approach 2:
The patent changes encoding parameters adaptively based on local depth characteristics. The weighting parameters based on distance, depth difference, and image difference are adjusted dynamically to preserve edge information in critical regions while allowing more compression in non-critical regions, thereby maintaining rendering quality at reduced bit rates.
3Manufacturing precision
If partition refinement is applied to preserve depth edges, then depth edge quality is improved, but encoding complexity increases
Solution Approach 1:
The encoding process segments the depth map into partitions and sub-partitions, allowing independent processing of different regions. This segmentation enables focused refinement on edge regions while maintaining simpler encoding in flat regions, balancing edge quality with encoding complexity through localized processing.
Solution Approach 2:
The patent applies partition refinement selectively rather than uniformly across the entire depth map. By focusing refinement actions on regions containing depth edges and using coarser partitioning in flat regions, the patent achieves good edge quality without the excessive complexity that would result from uniform fine-grained partitioning.
Data Source
AI summary
Several implementations relate, for example, to depth encoding and/or filtering for 3D video (3DV) coding formats. A sparse dyadic mode (308) for partitioning macroblocks (MBs) along edges in a depth map is provided as well as techniques for trilateral (or bilateral) filtering of depth maps that may include adaptive selection between filters sensitive to changes in video intensity and/or changes in depth. One implementation partitions a depth picture, and then refines the partitions based on a corresponding image picture. Another implementation filters a portion of a depth picture based on values for a range of pixels in the portion. For a given pixel in the portion that is being filtered, the filter weights a value of a particular pixel in the range by a weight that is based on one or more of location distance, depth difference, and image difference.