Parameterized Motion Model for Stereo Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression techniques for stereo video sequences, such as those used in virtual reality applications, are inefficient due to their reliance on purely translational motion models, which fail to accurately predict motion involving scaling, shearing, or rotational transformations, leading to suboptimal encoding and decoding performance.
Innovation Solution
The implementation of a parameterized motion model that uses warping parameters to predict motion within video blocks by determining block-level disparity and selecting plane normal candidates based on rate-distortion values, allowing for encoding and decoding that accounts for multiple motion types, including translational, rotational, and scaling motions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If purely translational motion models are used for video compression, then device complexity is reduced, but manufacturing precision (motion prediction accuracy) deteriorates
Solution Approach 1:
The patent transforms the motion model from a simple translational model to a parameterized model that includes warping parameters (scaling, shearing, rotation). This allows the system to adaptively select different motion transformation types based on the actual motion characteristics in the video sequence, thereby improving motion prediction accuracy without requiring a completely complex model structure.
Solution Approach 2:
The patent introduces dynamic selection of motion model types through plane normal candidates. Instead of using a fixed translational motion model, the system dynamically determines the appropriate motion transformation type (scaling, shearing, rotation, or translation) based on the calculated plane normals and rate-distortion values, making the motion model adaptable to different scene requirements.
2Manufacturing precision
If parameterized motion models with multiple transformation types are used, then manufacturing precision (motion prediction accuracy) is improved, but device complexity increases
Solution Approach 1:
The patent segments the motion prediction process into distinct stages: calculating plane normals, generating multiple plane normal candidates, evaluating rate-distortion values for each candidate, and selecting the optimal candidate. This segmentation allows the complex parameterized motion modeling to be broken down into manageable steps, reducing implementation complexity while maintaining high prediction accuracy.
Solution Approach 2:
The patent generates multiple plane normal candidates (excessive action) but only selects the one with the best rate-distortion value (partial action). This approach allows the system to explore multiple motion transformation possibilities without fully committing to all of them, thereby improving prediction accuracy while controlling the complexity of the final implementation.
3Manufacturing precision
If block-level disparity and plane normal candidate selection are performed, then manufacturing precision (encoding efficiency) is improved, but device complexity increases
Solution Approach 1:
The patent employs rate-distortion evaluation as a feedback mechanism to select the optimal plane normal candidate. By calculating the rate-distortion value for each candidate and selecting the one that minimizes this value, the system ensures that the increased encoding complexity leads to proportional improvements in decoding efficiency and video quality.
Data Source
AI summary
Video blocks of stereo or non-stereo video sequences are coded using a parameterized motion model. For example, encoding a current block of a stereo video sequence can include determining a block-level disparity between first and second frames and identifying plane normal candidates within the current block of the first frame based on the block-level disparity. One of the plane normal candidates is selected based on rate-distortion values, and warping parameters are determined for predicting motion within the current block using the selected plane normal candidate. The current block is then encoded using a reference block generated by applying the warping parameters. Decoding that encoded block can include receiving a bitstream representing an encoded stereo video sequence, determining warping parameters for predicting motion within the encoded block based on syntax elements encoded to the bitstream, and decoding encoded block using a reference block generated by applying the warping parameters.