360-Degree Video Face Boundary Handling for Accurate Block Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing VR systems face challenges in handling face discontinuities in 360-degree video coding, leading to inefficiencies in decoding and rendering due to the discontinuous nature of neighboring blocks in frame-packed pictures, which affects prediction and filtering processes.
Innovation Solution
The system determines the location of face discontinuities and adjusts decoding functions such as merge modes, intra-prediction, inter-prediction, and filtering processes based on the availability of neighboring blocks and reconstructed samples on the same side of the discontinuity, applying specific filters and prediction methods separately for each prediction unit across the discontinuity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional video coding is applied to 360-degree video, then coding simplicity is maintained, but decoding efficiency deteriorates due to face discontinuities causing incorrect prediction and filtering
Solution Approach 1:
The patent segments the 360-degree video into multiple face regions, each with its own discontinuity flags. By dividing the video content into manageable segments and treating each face independently with specific discontinuity markers, the system can apply targeted prediction and filtering operations that respect face boundaries, thereby improving decoding efficiency without requiring complete redesign of the coding framework
Solution Approach 2:
The patent performs preliminary identification and marking of face discontinuities during the encoding phase. Discontinuity flags are pre-calculated and embedded in the bitstream, allowing the decoder to proactively recognize and handle discontinuity regions without complex real-time analysis, thus improving decoding efficiency while maintaining relatively simple coding structure
2Measurement precision
If spatial neighbors from across face boundaries are used for prediction, then prediction accuracy may improve, but harmful artifacts are generated due to discontinuous face geometry
Solution Approach 1:
The patent applies different prediction strategies based on local conditions - specifically whether a block is adjacent to a face discontinuity. For blocks near discontinuities, the system uses discontinuity-aware prediction that restricts spatial neighbor selection to same-face blocks only, preventing artifact generation. For blocks away from discontinuities, conventional prediction can be used, maintaining prediction accuracy where applicable
Solution Approach 2:
The patent converts the potentially harmful effect of face discontinuities into a beneficial filtering mechanism. By explicitly marking discontinuity locations and using these markers to guide prediction operations, the system transforms what would be source of artifacts (discontinuous geometry) into a useful guide for making intelligent prediction decisions that avoid artifacts while maintaining accuracy
3Reliability
If standard filtering operations are applied across face boundaries, then filtering consistency is maintained, but decoding quality deteriorates due to incorrect sample usage across discontinuities
Solution Approach 1:
The patent modifies filtering operations to account for local discontinuity conditions. When a block is identified as being adjacent to a face discontinuity, the filtering process is adjusted to use only reconstructed samples from the same face, ensuring local consistency and preventing quality degradation. This localized adaptation maintains overall filtering consistency while improving decoding quality at critical boundary regions
Solution Approach 2:
The patent applies preliminary protective measures by identifying discontinuity regions and pre-adjusting filtering parameters before actual filtering operations. By anticipating potential quality issues at face boundaries and pre-configuring appropriate filtering behavior, the system prevents decoding quality deterioration before it occurs, while maintaining consistent filtering elsewhere
Data Source
AI summary
Systems, methods, and instrumentalities may be provided for discounting reconstructed samples and/or coding information from spatial neighbors across face discontinuities. Whether a current block is located at a face discontinuity may be determined. The face discontinuity may be a face boundary between two or more adjoining blocks that are not spherical neighbors. The coding availability of a neighboring block of the current block may be determined, e.g., based on whether the neighboring block is on the same side of the face discontinuity as the current block. For example, the neighboring block may be determined to be available for decoding the current block if it is on the same side of the face discontinuity as the current block, and unavailable if it is not on the same side of the face discontinuity. The neighboring block may be a spatial neighboring block or a temporal neighboring block.