Image Processing Apparatus Submatrix Derivation for Memory Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods require significant memory capacity for orthogonal and inverse orthogonal transforms due to the large size of lookup tables needed to hold transformation matrices, leading to increased memory requirements and circuit complexity.
Innovation Solution
Deriving a transformation matrix using a submatrix, which reduces the number of transformation matrices needed for orthogonal and inverse orthogonal transforms, thereby minimizing memory requirements by substituting similar transformation matrices and optimizing waveform characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple different orthogonal transforms are used for primary transform in adaptive multiple core transforms, then coding efficiency is improved, but memory capacity required for orthogonal transform increases
Solution Approach 1:
The transformation matrix is segmented into a submatrix and a complementary matrix. Only the submatrix is stored in the lookup table, while the complementary matrix is derived through matrix operations. This segmentation reduces the storage requirements while maintaining the capability to perform multiple transform types.
Solution Approach 2:
A derivation unit acts as an intermediary between the stored submatrix and the full transformation matrix. The derivation unit performs matrix operations to generate the complete transformation matrix from the submatrix, eliminating the need to store all transformation matrices directly.
2Adaptability or versatility
If multiple different orthogonal transforms are stored in lookup table, then transform versatility is improved, but device complexity increases
Solution Approach 1:
The system uses a universal derivation mechanism that can generate multiple different transformation matrices from a single submatrix through different matrix operations. This allows one storage structure to serve multiple transform functions, reducing device complexity while maintaining versatility.
Solution Approach 2:
Instead of storing multiple complete transformation matrices, the system stores a submatrix and creates copies/derivations of the full transformation matrix as needed through computational operations, reducing the physical storage requirements and circuit complexity.
Data Source
AI summary
The present disclosure relates to an image processing apparatus and an image processing method for enabling suppression of an increase in a memory capacity required for orthogonal transform and inverse orthogonal transform.A transformation matrix is derived using a submatrix configuring a part of the transformation matrix, a prediction residual of an image is orthogonally transformed using the derived transformation matrix, and coefficient data obtained by orthogonally transforming the prediction residual is encoded to generate a bit stream. The present disclosure can be applied to, for example, an image processing apparatus, an image encoding device, an image decoding device, or the like.