Adaptive Motion Vector Prediction in Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding methods, such as MPEG-4 H.264/AVC, face inefficiencies in predicting motion vectors for blocks, leading to suboptimal compression ratios due to fixed coding unit sizes and prediction methods that do not adapt well to varying image characteristics.
Innovation Solution
The method involves adaptive encoding and decoding of motion vectors using hierarchical coding units with variable sizes and depths, selecting between explicit and implicit prediction modes based on image characteristics and coding unit depth to optimize motion vector prediction and compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If fixed coding unit sizes and prediction methods are used, then device complexity is reduced, but compression ratio deteriorates due to inability to adapt to varying image characteristics
Solution Approach 1:
The coding unit is divided into multiple sub-coding units with different sizes (e.g., 16x16, 8x8, 4x4 blocks) that can be selectively applied to different regions of the image. This segmentation allows the system to adapt to varying image characteristics by choosing appropriate block sizes for different areas, thereby improving compression ratio without requiring a complete redesign of the coding structure.
Solution Approach 2:
The patent introduces dynamic selection of coding unit sizes and prediction modes based on image characteristics. The system can dynamically switch between different coding unit depths and prediction methods (e.g., intra-prediction, inter-prediction) depending on the local image content, enabling adaptability while maintaining a manageable device complexity through structured decision-making processes.
2Productivity
If motion vector prediction is performed using fixed methods, then processing speed is maintained, but compression ratio deteriorates due to suboptimal prediction accuracy
Solution Approach 1:
The patent changes the parameters of motion vector prediction by introducing multiple prediction modes (e.g., skip mode, direct mode, motion compensation modes) and allowing selection based on image characteristics. This enables the system to optimize prediction accuracy for different scenarios, improving compression ratio while managing complexity through parameter-based adaptation rather than structural overhaul.
3Productivity
If coding unit size is increased, then processing efficiency is improved, but prediction accuracy deteriorates due to inability to capture local image variations
Solution Approach 1:
By segmenting the image into multiple coding units of varying sizes, the system can process larger regions efficiently (maintaining processing efficiency) while also creating smaller sub-coding units in areas requiring high prediction accuracy. This hierarchical segmentation resolves the contradiction by allowing both large and small coding units to coexist in the same image processing pipeline.
Solution Approach 2:
The patent applies local quality by allowing different coding unit sizes and prediction methods to be used in different regions of the image based on local characteristics. Smooth regions can use larger coding units for processing efficiency, while regions with high detail or motion can use smaller coding units for prediction accuracy, thereby resolving the contradiction between processing efficiency and prediction accuracy.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Provided are methods and apparatuses for encoding and decoding a motion vector. The method of encoding the motion vector includes: selecting, as a mode of encoding information about a motion vector predictor of the current block, a first mode in which information indicating the motion vector predictor from among at least one motion vector predictor is encoded or a second mode in which information indicating generation of the motion vector predictor based on blocks or pixels included in a previously encoded area adjacent to the current block is encoded; determining the motion vector predictor of the current block according to the selected mode and encoding the information about the motion vector predictor of the current block; and encoding a difference vector between the motion vector of the current block and the motion vector predictor of the current block.