Global Motion Vector Coding With Control Point Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression technologies face inefficiencies in encoding and decoding due to complex relationships between video quality, data usage, encoding complexity, sensitivity to errors, and the need for improved compression efficiency, particularly in handling global and local motion within video frames.
Innovation Solution
The implementation of global motion vectors signaling in video codecs, utilizing control point motion vectors to derive affine motion parameters, and predicting these vectors from previous frames to reduce the bit rate and improve compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If motion compensation is used to predict video frames, then compression efficiency is improved, but encoding and decoding complexity increases
Solution Approach 1:
The patent segments motion vectors into control point motion vectors (CPMVs) at corner positions and residual motion vectors. By dividing the motion representation into control points and residuals, the system achieves better compression efficiency while managing decoding complexity through a structured approach to motion compensation.
Solution Approach 2:
The patent performs preliminary action by deriving affine motion parameters from control point motion vectors before actual motion compensation. This pre-computation of motion parameters from CPMVs allows the decoder to efficiently reconstruct motion fields without requiring complex real-time calculations during decoding.
2Measurement precision
If global motion vectors are encoded with high precision, then video quality is improved, but bit rate increases
Solution Approach 1:
The patent changes parameters by representing global motion through control point motion vectors at corner positions rather than encoding full motion vectors for all pixels. This parameter transformation allows high-precision motion representation at reduced bit rates by focusing computational effort on key control points that define the global motion field.
Solution Approach 2:
The patent applies partial action by encoding motion information only at control points (corner positions) rather than for every pixel. This partial encoding approach achieves sufficient video quality by capturing essential global motion characteristics at key locations, reducing overall bit rate requirements while maintaining acceptable reconstruction quality.
3Productivity
If control point motion vectors are predicted from previous frames, then compression efficiency is improved, but sensitivity to errors increases
Solution Approach 1:
The patent implements feedback by predicting control point motion vectors from previously decoded frames and using these predictions in the current frame's motion compensation. This feedback mechanism improves compression efficiency by exploiting temporal redundancy, while the residual encoding provides error protection to mitigate sensitivity to prediction errors.
Data Source
AI summary
A video encoder is configured to generate a bitstream for decoding by a compatible decoder which includes circuitry configured to receive a bitstream, extract a residual of a control point motion vector for a current frame and from the bitstream, and combine the residual of the control point motion vector with a prediction of the control point motion vector for the current frame.