Wedgelet Pattern Table Size Reduction in Depth Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The large size of the wedgelet pattern table in Depth Modeling Mode 1 (DMM1) for depth coding in three-dimensional and multi-view video systems leads to increased storage and bandwidth requirements, and reducing this size while maintaining coding efficiency is a challenge.
Innovation Solution
The proposed solution involves generating a reduced wedgelet pattern table by excluding certain samples from the starting and ending positions for adjacent-edge and opposite-edge partitions, and using this reduced table for encoding and decoding depth blocks, with specific constraints applied to ensure valid wedgelet pattern indices and shared table usage across different block sizes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a complete wedgelet pattern table is used for DMM1 depth coding, then coding efficiency is maintained, but storage space and bandwidth requirements increase significantly
Solution Approach 1:
The patent extracts and removes redundant wedgelet patterns from the complete pattern table. Specifically, it identifies and eliminates patterns that are not commonly used or that can be derived from other patterns, thereby reducing the table size while retaining the essential patterns needed for effective depth coding.
Solution Approach 2:
The patent makes the wedgelet pattern table universally applicable across different block sizes by using a single reduced table for multiple prediction unit sizes (4×4, 8×8, 16×16, 32×32). This is achieved by adapting the pattern selection and indexing mechanism to work efficiently across different scales, eliminating the need for separate complete tables for each block size.
2Quantity of substance
If a reduced wedgelet pattern table is used, then storage and bandwidth requirements decrease, but coding efficiency may be compromised
Solution Approach 1:
The patent changes the indexing parameters and selection criteria for wedgelet patterns. It introduces a modified indexing scheme that efficiently maps prediction units to the reduced pattern table, ensuring that the most suitable patterns are selected even with the reduced table size. This maintains coding efficiency by optimizing the parameter selection process.
Solution Approach 2:
The patent applies partial action by selecting only the most essential and frequently used wedgelet patterns for the reduced table, rather than including all possible patterns. This selective approach retains sufficient coding capability while significantly reducing storage requirements, achieving an optimal balance between table size and coding performance.
3Manufacturing precision
If different block sizes use separate complete wedgelet pattern tables, then each block size gets optimized patterns, but device complexity and memory requirements increase
Solution Approach 1:
The patent creates a universal reduced wedgelet pattern table that serves multiple block sizes (4×4, 8×8, 16×16, 32×32) simultaneously. This single table replaces what would otherwise require four separate complete tables, significantly reducing memory requirements and simplifying the table management architecture while maintaining effective pattern matching across all block sizes.
Solution Approach 2:
The patent introduces a new dimension of scalability by designing the reduced pattern table and indexing mechanism to adapt to different block sizes through parameter adjustment rather than requiring separate tables. This dimensional approach allows the same table structure to serve multiple purposes across different scales.
Data Source
AI summary
A method and apparatus of depth coding using depth modelling mode 1 (DMM1) are disclosed to reduce the wedgelet pattern table size. In one embodiment, a size-reduced wedgelet pattern for a reduced wedgelet pattern table is generated by excluding at least one non-corner adjacent-edge sample for adjacent-edge partition or at least one opposite-edge sample for opposite-edge partition from starting positions or from ending positions. The reduced wedgelet pattern table may also include at least one omitted wedgelet pattern in at least one wedgelet direction category. For the adjacent-edge partition, the starting positions and the ending positions may correspond to every other even non-corner adjacent-edge samples in a first and a second adjacent edges respectively. For the opposite-edge partition, the starting positions correspond to every other even opposite-edge samples in a first opposite edge and the ending positions include all opposite-edge samples in a second opposite edge.