Implicit Index Derivation for Video Slice Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques, such as those described in the Versatile Video Coding (VVC) working draft, face inefficiencies in bit allocation and signaling for tile, brick, and slice partitioning, particularly in deriving index values for the last rectangular slice of a coded picture.
Innovation Solution
The proposed method implicitly derives index values for the top-left and bottom-right bricks of the last slice based on the index values of the second-to-last slice and the total number of sub-picture units, eliminating the need for explicit signaling of these values and optimizing bit efficiency by using an implicit derivation rule.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If explicit signaling of index values for top-left and bottom-right bricks is performed, then decoding accuracy is ensured, but bitstream size increases
Solution Approach 1:
The patent enables the decoder to self-determine the index values of top-left and bottom-right bricks for the last rectangular slice by using the total number of bricks in the picture and the brick indexing rule, eliminating the need for explicit signaling of these values in the bitstream
Solution Approach 2:
The patent extracts and removes the redundant explicit signaling of index values for the last slice from the bitstream, keeping only the essential signaling for non-last slices while deriving the last slice indices implicitly from previously signaled information
2Loss of information
If explicit signaling of index values and difference values is performed, then complete information is provided, but bit efficiency decreases
Solution Approach 1:
The patent applies partial signaling by only explicitly signaling index values for non-last rectangular slices, while the last slice indices are partially derived from the total brick count and indexing rules, avoiding excessive signaling of redundant information
Data Source
AI summary
Methods and systems for decoding a picture. A method includes receiving a coded video stream including a picture partitioned into first sub-picture units, the first sub-picture units including one first sub-picture unit, and an additional first sub-picture unit including a first ordered second sub-picture unit, from among second sub-picture units of the additional first sub-picture unit, and a last ordered second sub-picture unit, from among the second sub-picture units of the additional first sub-picture unit. The method further including decoding the picture, the decoding including obtaining index values of the first ordered second sub-picture unit and the last ordered second sub-picture unit of the additional first sub-picture unit, without the coded video stream explicitly signaling any of the index values and a difference value between the index values of the first ordered second sub-picture unit and the last ordered second sub-picture unit to the at least one processor.


