Still Block Detector for Moving Image Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In moving image encoding, it is challenging to determine before frame processing begins whether discrete cosine transform (DCT) and quantization can be skipped for macroblocks, leading to unpredictable speedup in frame processing.
Innovation Solution
The technique involves dividing frames into blocks, selecting blocks with minimal movement amounts, and forcibly replacing motion vectors or DCT coefficients with zeros, thereby skipping unnecessary motion searches and computations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion search is performed for all macroblocks to determine whether DCT and quantization can be skipped, then encoding accuracy is improved, but frame processing time increases and speedup cannot be predicted beforehand
Solution Approach 1:
The frame is divided into multiple blocks, and a specific number S of blocks are selected to skip motion search and DCT/quantization operations. This segmentation allows the system to process different blocks differently, skipping computations for selected blocks while performing full processing for others, thereby reducing overall processing time while maintaining acceptable encoding accuracy.
Solution Approach 2:
Instead of performing motion search and DCT/quantization on all blocks, the invention applies these operations only to a portion of blocks (those not selected for skipping). By performing partial action on selected blocks and skipping actions on others, the system achieves speedup while maintaining sufficient encoding quality for the overall frame.
2Productivity
If motion search and DCT/quantization are skipped for more blocks, then frame processing speed is improved, but encoding quality may deteriorate
Solution Approach 1:
The invention changes the parameter of block processing by selecting S blocks to skip computations. By adjusting the number S of blocks to skip, the system can control the balance between processing speed and encoding quality. This parameter change allows flexible optimization based on specific encoding requirements and performance targets.
3Manufacturing precision
If the number of blocks to skip is determined dynamically based on actual difference image analysis, then encoding quality is maintained, but the complexity of determining skip candidates increases
Solution Approach 1:
The invention performs preliminary selection of S blocks to skip before actual encoding processing. By determining skip candidates in advance based on simple criteria rather than performing full difference image analysis on all blocks, the system reduces computational complexity while still achieving acceptable encoding quality. The preliminary action of selecting which blocks to skip simplifies the overall processing pipeline.
Data Source
AI summary
A still block detector is provided at a stage before a moving image encoder. The still block detector divides a frame to be processed into a plurality of blocks, sorts the blocks in the frame in ascending order of movement amount, and sets the first one or more blocks the number of which is specified by a “corresponding block setting number” input from the outside of an encoding device, to “still blocks.” For the still blocks, the moving image encoder performs an inter-encoding process with respect to a moving image by forcibly setting motion vectors to zeros without performing a motion search or by forcibly setting DCT coefficients to zeros without performing DCT or quantization.