Resampling Bit Depth Constraint for Video Coding Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In scalable video coding, the intermediate data from the resampling process often has a bit depth that exceeds the maximum allowed by certain computational instruction sets, leading to increased computational complexity and rounding errors, particularly in vertical resampling steps after horizontal resampling.
Innovation Solution
Constraining the bit depth of intermediate output from the initial resampling process to a specified number, such as 16 bits, by right shifting the output based on the input bit depth, to mitigate these issues and allow for more efficient coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the intermediate data from resampling process is kept at high bit depth, then measurement precision is improved, but device complexity increases due to exceeding maximum allowed bit depth of computational instruction sets
Solution Approach 1:
The patent applies preliminary action by performing the bit depth constraint operation immediately after the first horizontal resampling step, before the second vertical resampling step. This preliminary constraint prevents the intermediate data from exceeding the maximum allowed bit depth, thereby avoiding increased computational complexity in subsequent processing steps while maintaining sufficient precision for the resampling operation.
2Measurement precision
If the intermediate data from resampling process is kept at high bit depth, then measurement precision is improved, but loss of time increases due to increased rounding errors requiring correction
Solution Approach 1:
The patent performs the bit depth constraint operation in advance, immediately after the first resampling step, to prevent rounding errors from accumulating during subsequent processing. By establishing the appropriate bit depth early in the process, the patent eliminates the need for time-consuming rounding error corrections that would be required if high bit depth data were maintained throughout the entire resampling process.
3Device complexity
If the intermediate data from resampling process is constrained to lower bit depth, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent applies the bit depth constraint at the optimal point in the processing sequence - immediately after the first horizontal resampling step - rather than applying it too early or too late. This timing ensures that the data has sufficient precision to complete the necessary resampling calculations, while the constraint is applied before the data would exceed the maximum allowed bit depth, thus maintaining an appropriate balance between precision and computational complexity.
Data Source
AI summary
An apparatus for coding video information according to certain aspects includes a memory and a processor. The memory unit is configured to store video information. The processor is configured to: obtain reference layer video information; upsample the reference layer video information in a first dimension to generate an intermediate output; constrain the intermediate output to a predetermined bit depth; and upsample the constrained intermediate output in a second dimension, wherein the second dimension is orthogonal to the first dimension.