Projection-Aware 360 Image Decoding for High-Data VR Video
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems struggle with the massive data generated from processing multi-view 360-degree images for virtual and augmented reality, leading to insufficient performance in encoding and decoding high-resolution 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 various projection formats, such as Equi-Rectangular, CubeMap, OctaHedron, and IcoSahedral, while optimizing image expansion and motion vector prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multi-view 360-degree images are processed for virtual reality and augmented reality, then the quality and realism of media service is improved, but the amount of data generated increases massively
Solution Approach 1:
The patent segments the 360-degree image into multiple view images captured by different cameras, allowing independent processing and encoding of each view. This segmentation enables efficient data management by treating each view as a separate entity that can be encoded and transmitted independently, reducing the overall data burden while maintaining quality.
Solution Approach 2:
The patent uses prediction techniques where reference pictures (copies of previously decoded images) are utilized to predict current image content. By copying and reusing reference picture data through motion compensation and prediction algorithms, the system reduces redundant data transmission while preserving image quality.
2Area of stationary object
If the amount of data for 360-degree images is processed, then comprehensive coverage is improved, but the performance of image processing system becomes insufficient
Solution Approach 1:
The patent divides the comprehensive 360-degree coverage into multiple discrete camera views, each processed independently through dedicated encoding pipelines. This segmentation allows parallel processing of different view segments, improving overall system productivity while maintaining complete spherical coverage.
Solution Approach 2:
The patent performs prediction and estimation operations using reference pictures before actual decoding. By preliminarily predicting image content based on motion vectors and reference frames, the system reduces the computational burden during final decoding, improving processing performance while maintaining full coverage.
3Measurement precision
If high-resolution images are encoded and decoded, then image quality is improved, but the processing time and system requirements increase
Solution Approach 1:
The patent extensively uses reference picture copying and motion compensation to predict high-resolution image content. By copying motion vectors and prediction data from reference frames, the system achieves high-resolution decoding with reduced processing time, as the actual pixel data needs to be processed less intensively when prediction is accurate.
Solution Approach 2:
The patent dynamically adjusts encoding parameters such as quantization levels, transformation block sizes, and prediction modes based on image content characteristics. This adaptive parameter adjustment optimizes the balance between resolution quality and processing time, allowing high-resolution output while minimizing unnecessary computational overhead.
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.