Adaptive Motion Vector Accuracy for Content-Aware Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards, such as MPEG-4 AVC, use uniform motion vector accuracy without considering the relationship between motion information and video content, leading to suboptimal rate-distortion cost and inefficient coding.
Innovation Solution
Adaptive motion vector accuracy schemes are employed based on non-rate-distortion criteria, including partition size, motion vector directions, quantization parameters, and video content, to selectively adjust motion vector accuracy for improved coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If uniform motion vector accuracy (quarter-pel accuracy) is used for all blocks, then motion compensation performance is improved, but coding cost and bit rate increase
Solution Approach 1:
The patent applies local quality by differentiating motion vector accuracy requirements across different block types. Specifically, quarter-pel accuracy is applied only to inter-block motion vectors in 16x16 and 8x8 blocks, while half-pel accuracy is used for intra-block motion vectors and other cases. This localized application of high accuracy only where needed optimizes the balance between motion compensation performance and coding cost.
Solution Approach 2:
The patent changes the motion vector accuracy parameter dynamically based on block size and type. The accuracy parameter is not fixed but adapts according to the specific requirements of different blocks, allowing the system to use higher precision where motion compensation is most critical and lower precision where it is sufficient, thereby reducing overall bit rate.
2Productivity
If quarter-pel accuracy motion vectors are used everywhere, then coding gains are achieved over past standards, but the cost to code higher accuracy increases
Solution Approach 1:
The patent implements local quality by applying quarter-pel accuracy only to specific block types (inter-block motion vectors in 16x16 and 8x8 blocks) while using half-pel accuracy for other cases. This selective application maintains coding efficiency gains where needed while reducing coding cost elsewhere, avoiding the unnecessary computational burden of applying high accuracy uniformly.
Solution Approach 2:
The patent applies partial action by using quarter-pel accuracy only partially (for specific block types) rather than excessively (for all blocks). This partial application provides sufficient coding gains for the critical cases while avoiding the excessive computational cost that would result from applying the higher accuracy universally.
3Measurement precision
If motion vector accuracy is increased to represent all types of motion accurately, then motion compensation quality improves, but bandwidth and memory requirements increase
Solution Approach 1:
The patent applies local quality by using quarter-pel accuracy only for inter-block motion vectors in large blocks where motion representation accuracy is most critical, while using half-pel accuracy for intra-block vectors and other cases. This localized high precision reduces the overall bandwidth requirement compared to applying quarter-pel accuracy universally.
Solution Approach 2:
The patent dynamically changes the motion vector accuracy parameter based on block characteristics. By adapting the precision parameter to match the actual motion representation needs of different blocks, the system achieves sufficient measurement precision where required while minimizing bandwidth consumption overall.
Data Source
AI summary
Methods and apparatus are provided for adaptive coding of motion information. An apparatus includes an encoder (100) for encoding at least a block in a picture using a motion vector. An adaptive motion vector accuracy scheme is used to select an accuracy of the motion vector used to encode the block. Selection criteria for selecting the accuracy for the motion vector include non-rate-distortion-based criteria.