Superblock Motion Vector Coding for Video Compression Bit Rate Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression techniques face high variability in bit rate due to motion vectors, requiring excessive reserved bandwidth for accommodating occasional increases, which is often wasted.
Innovation Solution
Partitioning video frames into superblocks of macroblocks and coding motion vectors to achieve fixed-rate motion vectors by determining optimal block sizes and resolutions that minimize residual complexity within pre-defined bit constraints, ensuring consistent bit allocation across superblocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion vectors are encoded with high accuracy (e.g., 1/8 pixel resolution), then prediction precision is improved, but the number of bits required to encode motion vectors increases, requiring greater bandwidth
Solution Approach 1:
The video frame is divided into superblocks containing multiple macroblocks. Motion vectors are encoded at different precision levels for different macroblocks within the same superblock, allowing high precision where needed and lower precision where acceptable, thereby reducing overall bit rate while maintaining prediction quality.
Solution Approach 2:
Different macroblocks within a superblock are assigned different motion vector precision levels based on local characteristics. Macroblocks with significant motion or complex patterns receive higher precision (e.g., 1/4 or 1/8 pixel), while those with simple or static content use lower precision (e.g., 1 or 1/2 pixel), optimizing the trade-off between accuracy and bandwidth.
2Reliability
If bandwidth is reserved to accommodate peak bit rate requirements, then reliability of transmission is improved, but bandwidth efficiency deteriorates due to wasted reserved capacity during low-activity periods
Solution Approach 1:
The system dynamically adjusts motion vector precision and block partitioning strategies based on actual content characteristics and motion activity. This adaptability causes the bit rate to closely follow the actual information content rather than requiring static over-provisioning, improving bandwidth efficiency while maintaining reliable transmission through rate control mechanisms.
Solution Approach 2:
The encoder changes multiple parameters including block size, motion vector precision, and partitioning granularity based on local content analysis. These parameter adjustments allow the system to match bit rate to actual complexity requirements, eliminating the need for excessive bandwidth reservation while ensuring reliable delivery of essential video information.
3Measurement precision
If frame is partitioned into smaller block sizes for motion estimation, then motion vector accuracy is improved, but the complexity of the encoding process increases
Solution Approach 1:
The frame is segmented into superblocks containing multiple macroblocks, which are further divided into smaller blocks only where necessary. This hierarchical segmentation allows the encoder to perform motion estimation at multiple levels, reducing overall computational complexity compared to dividing the entire frame into small blocks, while still achieving high accuracy where needed.
Solution Approach 2:
Instead of applying fine-grained block partitioning uniformly across the entire frame, the encoder applies smaller block sizes only to specific macroblocks where motion complexity or prediction accuracy requirements demand it. This partial application of fine partitioning significantly reduces encoding complexity while maintaining motion vector accuracy for critical regions.
Data Source
AI summary
Systems, devices, and methods of performing inter-frame prediction in video compression. A method includes: partitioning a video frame into superblocks, each superblock includes a plurality of macroblocks; determining an upper boundary on attributes that are associated with residual blocks of the plurality of macroblocks, based on pre-defined constraints on the total number of bits allocated for motion vector descriptions in a superblock; for each macroblock in a superblock, determining a partition size and a motion vector resolution which yield, for all macroblocks in the superblock, a motion vector description and a corresponding residual with attributes less than or equal to the upper boundary.