Offset Temporal Motion Vector Predictor for HEVC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current High Efficiency Video Coding (HEVC) systems face inefficiencies in motion vector prediction, particularly when there is significant object motion between frames, leading to suboptimal prediction accuracy and increased overhead bits for motion vector signaling.
Innovation Solution
The method introduces an offset to the temporal motion vector predictor (TMVP) by adding a coordinate offset derived from spatially neighboring blocks, allowing for improved prediction accuracy and reduced overhead bits through explicit or implicit signaling, and enabling the use of a shared offset for multiple prediction blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional temporal motion vector prediction is used, then overhead bits for motion vector signaling are reduced, but prediction accuracy deteriorates when there is significant object motion between frames
Solution Approach 1:
The patent introduces an offset parameter (dx, dy) to modify the traditional TMVP coordinate calculation. Instead of using the direct temporal collocated block coordinates, the patent adjusts the coordinates by adding an offset derived from spatial neighboring blocks, thereby improving prediction accuracy without significantly increasing overhead bits.
Solution Approach 2:
The patent uses spatial neighboring blocks as an intermediary to bridge the gap between temporal motion prediction and actual object motion. The motion vectors from spatial neighbors serve as a mediator to calculate the offset, which then corrects the temporal motion vector predictor to better match the actual motion.
2Measurement precision
If offset TMVP is introduced to improve prediction accuracy, then motion vector prediction accuracy improves, but device complexity increases
Solution Approach 1:
The patent applies offset TMVP selectively rather than universally. The offset calculation is performed only when beneficial, and the offset values are derived from a limited set of spatial neighboring blocks (typically left and above blocks), avoiding excessive computation while still achieving improved accuracy where needed.
Solution Approach 2:
The patent calculates offsets locally for each prediction block based on its specific spatial neighbors. Each block gets a customized offset based on the motion characteristics of its adjacent blocks, rather than applying a global offset to all blocks, thereby improving accuracy locally without uniformly increasing complexity across the entire video stream.
Data Source
AI summary
An improved method for temporal motion vector prediction for inter block HEVC is provided that relies on a block translational model. The method adds an offset to a temporal motion vector predictor (TMVP) to improve prediction accuracy. The method first designates a current prediction block as an area for motion compensation where all the pixels inside the prediction block perform identical translation temporally using motion vectors MVs. A coordinate offset is then derived for a current prediction block from the MVs of its spatially neighboring blocks. The offset TMVP is then defined for the current prediction block as the MV of an offset block which is in the geometrical location of the current prediction block coordinate plus the coordinate offset in a specified temporal reference picture. The offset TMVP is then used to code MVs.