Mesh Motion Vector Prediction for Efficient Vertex Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The encoding efficiency of motion vectors in existing mesh encoding technologies is low.
Innovation Solution
A mesh decoding device and method that includes an inter decoding unit to decode vertex coordinates by adding reference frame coordinates and motion vectors, utilizing a motion vector residual decoding unit, motion vector prediction unit, and motion vector calculation unit to optimize motion vector usage, thereby improving encoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If motion vectors are encoded using existing mesh encoding technologies, then the mesh can be represented with motion information, but the encoding efficiency is low and the data size is large
Solution Approach 1:
The patent applies preliminary action by performing motion vector prediction before encoding the actual motion vectors. The prediction unit calculates predicted motion vectors using neighboring vertex motion vectors, and only the residuals (differences between actual and predicted values) are encoded. This preliminary prediction step reduces the amount of data that needs to be encoded, directly addressing the contradiction between data size and encoding efficiency
Solution Approach 2:
The patent changes the parameter representation by encoding motion vector residuals instead of absolute motion vector values. By transforming the encoding target from original motion vectors to their differences (residuals) after prediction, the data requires fewer bits to represent, thereby reducing data size while maintaining encoding efficiency
2Measurement precision
If all neighboring vertex motion vectors are used for prediction, then prediction accuracy improves, but computational complexity and processing time increase
Solution Approach 1:
The patent applies local quality by making the prediction process adaptive to local characteristics. Different prediction modes are selected based on the local geometry and motion characteristics of the mesh. The prediction unit evaluates multiple candidate motion vectors from neighboring vertices and selects the best match, allowing high prediction accuracy in regions with simple motion while reducing complexity in more complex regions
Solution Approach 2:
The patent introduces dynamics by making the prediction process flexible and adaptive rather than static. The system dynamically selects prediction modes and adjusts the number of neighboring vertices used for prediction based on local conditions. This dynamic approach allows the system to optimize between accuracy and complexity in real-time during encoding, addressing the contradiction between prediction accuracy and processing complexity
3Manufacturing precision
If motion vector residuals are calculated and added to predicted values, then accurate motion vectors are obtained, but additional computational steps are required
Solution Approach 1:
The patent applies segmentation by dividing the motion vector decoding process into distinct segments: prediction segment and residual addition segment. The motion vector calculation unit separately computes the predicted motion vector from neighboring vertices and then adds the decoded residual to obtain the final motion vector. This segmentation allows each step to be optimized independently while maintaining overall accuracy, addressing the contradiction between decoding accuracy and process complexity
Data Source
AI summary
In a mesh decoding device 200, in a mode 1, a motion vector calculation unit 202E4 outputs a motion vector of a vertex to be decoded by adding a motion vector residual and a motion vector predicted value, and in a mode 0, the motion vector calculation unit 202E4 outputs the motion vector residual as the motion vector of the vertex to be decoded.