Warping Signaling for Video Stabilization and Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently encoding and decoding video data, particularly in handling complex intra prediction modes and motion vectors, leading to suboptimal compression efficiency and increased data requirements.
Innovation Solution
The implementation of warping-based decoded picture resampling (WDPR) using supplemental enhancement information (SEI) messages to apply warping processes for image stabilization, allowing for more efficient encoding and decoding by adjusting pixel positions based on corner vectors and flags indicating the application of warping processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional video coding techniques are used, then compression is achieved, but compression efficiency is suboptimal and data requirements remain high
Solution Approach 1:
The patent applies warping processes to decoded pictures before they are used for further encoding. By pre-warping pictures to stabilize motion and correct geometric distortions, the system prepares the data in advance to reduce redundancy more effectively, thereby improving compression efficiency without increasing data requirements
Solution Approach 2:
The patent modifies picture parameters through warping operations, including geometric transformation, motion compensation, and resampling. These parameter changes enable better alignment of corresponding regions across frames, reducing motion blur and improving prediction accuracy, which directly enhances compression efficiency
2Measurement precision
If complex intra prediction modes and motion vectors are handled, then encoding accuracy improves, but processing complexity increases
Solution Approach 1:
The patent divides the picture into multiple blocks and applies different warping operations to each block based on its specific characteristics. By segmenting the processing task, the system can handle complex intra prediction modes and motion vectors in a manageable manner, maintaining encoding accuracy while reducing overall processing complexity through localized operations
3Manufacturing precision
If warping processes are applied to stabilize images, then image quality improves, but processing time increases
Solution Approach 1:
The patent integrates warping operations into the existing video encoding pipeline, allowing warping to occur continuously as pictures are decoded and processed. By making the warping process part of the standard encoding flow rather than a separate post-processing step, the system maintains image quality improvement while minimizing additional processing time through seamless integration
Data Source
AI summary
Aspects of the disclosure provide a method, an apparatus, and non-transitory computer-readable storage medium for video decoding. The apparatus includes processing circuitry that is configured to decode a current picture based on a bitstream. The processing circuitry can determine, from a supplemental enhancement information (SEI) message, a first flag indicating whether a warping process is applied to the current decoded picture. Based on the first flag indicating that the warping process is applied to the current decoded picture, the processing circuitry can determine warping information of the warping process based on the SEI message. The processing circuitry can determine a warped picture from the current decoded picture based on the warping information.