Image Decoding Apparatus Adaptive Quantization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image encoding techniques, such as H.264, face inefficiencies in intra prediction mode encoding and require a large number of bits to transmit adaptive quantization parameters, leading to low encoding and decoding efficiency.
Innovation Solution
An image decoding device that adaptively adjusts the quantization parameter of a current block based on the size of a coding unit by using a quantization parameter predictor derived from the left and top coding units, reducing the number of bits required for transmission and enhancing compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If adaptive quantization parameters are used for different coding unit sizes, then image quality and compression efficiency are improved, but the number of bits required for transmitting quantization parameters increases
Solution Approach 1:
The patent applies preliminary action by pre-defining a correspondence relationship between coding unit sizes and quantization parameter sets before encoding. The encoder selects and transmits only an index identifying the appropriate quantization parameter set rather than transmitting all quantization parameters, thereby reducing transmission bits while maintaining adaptive quantization capability for different coding unit sizes
Solution Approach 2:
The patent implements local quality by allowing different quantization parameter sets to be applied to different coding unit sizes (e.g., 16×16, 32×32, 64×64 blocks). Each coding unit size can have its own optimized quantization parameters tailored to its specific characteristics, improving compression efficiency while using index-based transmission to minimize bit overhead
2Measurement precision
If intra prediction mode is encoded for every current block, then prediction accuracy is improved, but encoding complexity and bit rate increase
Solution Approach 1:
The patent merges the encoding of intra prediction modes by grouping current blocks that share the same optimal intra prediction mode into the same coding unit. Instead of encoding individual mode decisions for each block, the encoder determines a single intra prediction mode for the entire coding unit, reducing encoding complexity and bit rate while maintaining prediction accuracy through unified mode selection
Solution Approach 2:
The patent applies segmentation by dividing the picture into coding units of various sizes and allowing different intra prediction mode encoding strategies for different coding unit sizes. Larger coding units use unified mode encoding to reduce complexity, while smaller units can use more detailed mode encoding when beneficial, creating a segmented approach that balances accuracy and complexity
Data Source
AI summary
An image encoding method can include determining an intra prediction mode of a current prediction unit; constructing a mode group including three intra prediction modes using valid intra prediction modes of left and top blocks of the current prediction unit; determining a mode group indicator and a prediction mode index of the current prediction unit; and encoding the mode group indicator and the prediction mode index, in which the mode group indicator indicates whether or not the intra prediction mode of the current prediction unit belongs to the mode group, when the mode group indicator indicates that the intra prediction mode of the current prediction unit belongs to the mode group, the prediction mode index specifies the intra prediction mode of the current prediction unit in the mode group, when the mode group indicator indicates that the intra prediction mode of the current prediction unit does not belong to the mode group, the prediction mode index specifies the intra prediction mode of the current prediction unit among intra prediction modes besides the three intra prediction modes in the mode group.