Direction-Scaled Inter Prediction for 360-Degree Image Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems struggle with the massive data requirements of 360-degree images for virtual and augmented reality, necessitating improved performance in image encoding and decoding, particularly for high-resolution and ultra-high-definition images.
Innovation Solution
A method for encoding and decoding 360-degree images that includes generating a predicted image using syntax information, combining it with a residual image, and reconstructing the image in specific projection formats, such as Equi-Rectangular, CubeMap, OctaHedron, and IcoSahedral, while utilizing motion vector candidates and reference pictures for enhanced compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If 360-degree images are processed for virtual reality and augmented reality, then realistic media service quality is improved, but the amount of data generated increases massively
Solution Approach 1:
The patent segments the 360-degree image into multiple projection formats (ERP, CMP, OHP, ISP) and processes different regions with different encoding parameters. By dividing the image into partitioning units and applying region-wise packing, the system reduces the overall data quantity while maintaining service quality through selective processing of different image regions.
Solution Approach 2:
The patent changes encoding parameters based on projection format type. Different quantization parameters, transformation parameters, and motion compensation parameters are applied according to the specific projection format (ERP, CMP, OHP, or ISP), optimizing compression efficiency for each format while preserving necessary image quality for VR/AR services.
2Manufacturing precision
If high-resolution and ultra-high-definition images are processed, then image quality is improved, but the performance of the image processing system becomes insufficient
Solution Approach 1:
The patent applies different processing qualities to different regions of the image based on projection format. Critical regions requiring high quality (such as face directions in VR) are processed with higher precision, while less critical regions use lower precision processing, thereby maintaining overall image quality while improving system performance.
Solution Approach 2:
The patent performs preliminary projection format selection and image expansion before main encoding. By pre-processing the image to determine the appropriate projection format and expanding the image accordingly, the system prepares the data in an optimized state for subsequent encoding, reducing the computational burden during the main processing stage.
3Ease of manufacture
If conventional image encoding methods are used, then processing simplicity is maintained, but compression performance for 360-degree images is insufficient
Solution Approach 1:
The patent creates a universal encoding framework that handles multiple projection formats (ERP, CMP, OHP, ISP) through a single integrated process. The encoding apparatus universally supports all these formats by selecting appropriate parameters based on the input format, maintaining processing simplicity while achieving good compression performance across different 360-degree image types.
Solution Approach 2:
The patent introduces projection format information as an intermediary element that mediates between the raw 360-degree image and the encoding process. This intermediary carries essential information about the image structure, enabling the encoder to automatically adapt its parameters without complex manual configuration, thus maintaining simplicity while improving compression.
Data Source
AI summary
A method of decoding an image, includes obtaining at least one offset for a picture, deriving a variable for scaling for the picture based on the at least one offset, and performing inter prediction based on the variable for scaling for the picture. The at least one offset is defined with a direction of scaling.