Screen Content Video MVD Coding with Adaptive Exponential Golomb Order
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video coding methods for screen content, such as those using the HEVC standard, face inefficiencies in encoding motion vector differences (MVDs) due to unpredictable motion vectors and the exponential increase in bits required for Exponential Golomb coding, leading to decreased coding efficiency.
Innovation Solution
Implementing a method that determines the absolute value of MVDs and encodes them using higher-order Exponential Golomb coding when the value exceeds a threshold, along with unified coding techniques for both MVDs and intra block copy vector differences (BVDs) to improve coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If Exponential Golomb coding is used for MVD encoding, then coding efficiency is improved for small MVD values, but the number of bits increases exponentially for large MVD values
Solution Approach 1:
The patent changes the coding parameter (Exponential Golomb coding order) based on the MVD value. When the absolute MVD value exceeds a threshold, higher-order Exponential Golomb coding (order > 1) is used instead of the conventional order 1, adapting the coding strategy to the actual data characteristics to reduce bit consumption for large MVD values
2Adaptability or versatility
If conventional MVD coding methods are used for screen content, then compatibility with existing standards is maintained, but coding efficiency decreases due to unpredictable motion vectors
Solution Approach 1:
The patent applies different coding strategies to different types of video content. For screen content video data where motion vectors are unpredictable, higher-order Exponential Golomb coding is specifically applied, while maintaining conventional coding for other content types, thus optimizing for local content characteristics
3Productivity
If separate coding methods are used for MVDs and BVDs, then specific optimization for each vector type is achieved, but device complexity increases
Solution Approach 1:
The patent merges the coding methodology for MVDs and BVDs by applying the same higher-order Exponential Golomb coding approach to both motion vector differences and intra block copy vector differences. This unified approach maintains coding efficiency optimization while reducing the complexity of maintaining separate coding paths
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
A method for motion vector difference (MVD) coding of screen content video data is disclosed. In one aspect, the method includes determining an MVD between a predicted motion vector and a current motion vector and generating a binary string comprising n bins via binarizing the MVD. The method further includes determining whether an absolute value of the MVD is greater than a threshold value and encoding a subset of the n bins via an exponential Golomb code having an order that is greater than one in response to the absolute value of the MVD being greater than the threshold value.