Dynamic Block Grouping for Video Encoding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional block-based video encoding methods, such as MPEG and H.264/AVC, have limitations in accurately encoding and decoding images due to fixed macroblock sizes and inefficient prediction modes, leading to suboptimal compression efficiency and increased computational complexity.
Innovation Solution
The proposed method classifies image data into basic blocks, groups, and subgroups, determining an encoding mode for each group based on error rates generated by different prediction modes, allowing for dynamic data processing units and adaptive encoding strategies, including frequency transformation and motion estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed macroblock sizes are used in conventional block-based video encoding, then the encoding structure is simple and standardized, but the encoding accuracy and compression efficiency deteriorate
Solution Approach 1:
The patent applies dynamics by transitioning from fixed macroblock sizes to variable block sizes that can be dynamically determined based on image characteristics. The encoder analyzes the input image and determines optimal block sizes for different regions, allowing the encoding structure to adapt to varying content complexity rather than using a rigid fixed-size approach throughout the entire image.
Solution Approach 2:
The patent implements local quality by allowing different regions of the image to be encoded with different block sizes according to their specific characteristics. Complex regions use smaller blocks for better detail representation, while simple regions use larger blocks for efficiency, creating a non-uniform encoding structure that optimizes local encoding quality rather than applying a uniform fixed-size approach everywhere.
2Power
If fixed macroblock sizes are used, then the encoding process is computationally efficient, but the compression efficiency deteriorates
Solution Approach 1:
The patent applies dynamics by implementing a two-stage encoding process where the block size is dynamically determined in the first stage based on image analysis, and then fixed-size encoding is applied in the second stage using the determined block sizes. This dynamic determination allows the system to adapt to different image characteristics and optimize compression efficiency without excessive computational complexity throughout the entire encoding process.
Solution Approach 2:
The patent implements preliminary action by performing image analysis and block size determination before the actual encoding process. The encoder first analyzes the input image to determine optimal block sizes for different regions, and then uses these pre-determined sizes to guide the encoding process, reducing the computational burden during encoding while improving compression efficiency.
3Measurement precision
If multiple prediction modes are tested to optimize encoding, then the encoding accuracy improves, but the computational complexity increases
Solution Approach 1:
The patent applies preliminary action by determining the optimal block size for each macroblock before testing different prediction modes. This preliminary block size determination reduces the search space for prediction mode testing, as the encoder only needs to test prediction modes within the constraints of the pre-determined block sizes rather than evaluating all possible block size and prediction mode combinations, thereby reducing computational complexity while maintaining encoding accuracy.
Solution Approach 2:
The patent implements local quality by determining different block sizes for different macroblocks based on their specific characteristics, and then testing prediction modes locally for each macroblock with its optimized block size. This localized optimization approach allows the system to achieve high encoding accuracy for each region without the computational burden of testing all prediction modes across all possible block sizes globally.
Data Source
AI summary
An image encoding method is provided, in which image data divided into basic blocks is classified in units of groups and subgroups, wherein each group comprises at least one basic block and each subgroup comprises at least one basic block and is included in each group; an encoding mode for a predetermined group is determined in order to encode the predetermined group, wherein the encoding mode represents a mode for encoding data included in the predetermined group in units of one data processing unit selected from a group, a subgroup, and a basic block; and the data of the predetermined group is encoded according to the determined encoding mode. Detailed operations in the image encoding method are performed in consideration of the encoding mode of the group.