Image Codec Secondary Transform Signaling for Lower Decoding Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The secondary transform identifier in image encoding is signaled after each transform unit in a coding unit, leading to increased decoding processing load due to the need to complete decoding of all transform units and the identifier before starting inverse quantization and inverse transform processing.
Innovation Solution
The secondary transform identifier is set to be performed only for block sizes equal to or less than a predetermined threshold, and is signaled in a portion other than the footer of the encoded block, such as the header, to reduce decoding processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the secondary transform identifier is signaled after each transform unit in a coding unit, then the encoding flexibility is improved, but the decoding processing load increases
Solution Approach 1:
The patent divides the coding unit into multiple transform units and applies secondary transform selectively to each transform unit based on its own characteristics. This segmentation allows the secondary transform identifier to be determined and signaled for each transform unit independently, improving encoding flexibility while managing decoding complexity through localized processing.
Solution Approach 2:
The patent determines the secondary transform identifier for each transform unit before signaling it in the bitstream. This preliminary determination allows the decoder to prepare for inverse quantization and inverse transform processing in advance, reducing the processing load during actual decoding by having the transformation parameters ready before processing begins.
2Manufacturing precision
If the secondary transform identifier is determined depending on the total number of nonzero transform coefficients in the CU, then the encoding accuracy is improved, but the decoding processing delay increases
Solution Approach 1:
The patent performs preliminary counting of nonzero transform coefficients for each transform unit and determines the secondary transform identifier before the actual decoding process. This allows the decoder to have all necessary transformation parameters prepared in advance, enabling parallel processing of inverse quantization and inverse transform without waiting for complete CU decoding, thus reducing processing delay while maintaining encoding accuracy.
Solution Approach 2:
The patent segments the coding unit into multiple transform units and processes each independently with its own secondary transform identifier. This segmentation allows different parts of the CU to be decoded in parallel once their identifiers are determined, reducing overall decoding processing delay while maintaining the accuracy benefits of coefficient-based determination.
3Manufacturing precision
If inverse quantization and inverse transform processing wait for completion of all TU decoding and secondary transform identifier decoding, then the processing accuracy is improved, but the productivity decreases
Solution Approach 1:
The patent determines and signals the secondary transform identifier for each transform unit before inverse quantization and inverse transform processing begin. This preliminary action allows the decoder to have all transformation parameters ready in advance, enabling immediate processing without waiting for complete CU or TU decoding, thus improving productivity while maintaining processing accuracy through proper parameter availability.
Solution Approach 2:
The patent implements a dynamic decoding process where the secondary transform identifier for each transform unit is determined and made available at the appropriate time before its corresponding inverse processing. This dynamic approach allows different transform units to be processed at different rates based on their individual requirements, optimizing both accuracy and throughput by adapting the processing flow to the actual data characteristics.
Data Source
AI summary
The present disclosure relates to an image processing device and method capable of suppressing an increase in load of decoding processing. A secondary transform identifier is set such that secondary transform is performed only in a case where information regarding a block size is equal to or less than a predetermined threshold value, secondary transform is performed on coefficient data derived from image data on the basis of the secondary transform identifier set, and the secondary transform identifier set is encoded and a bitstream is generated. The present disclosure can be applied, for example, to an image processing device, an image encoding device, an image decoding device, an information processing device, an electronic device, an image processing method, an information processing method, and the like.