Projection-Aware 360-Degree Image Reconstruction for Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing image processing systems face challenges in efficiently handling the massive data generated for 360-degree images used in virtual and augmented reality, requiring improved performance in image encoding and decoding methods.
Innovation Solution
A method for decoding 360-degree images involves generating a predicted image using syntax information, combining it with a residual image, and reconstructing the image in a specific projection format, such as Equi-Rectangular, CubeMap, or OctaHedron, while considering region-wise packing and image expansion based on partitioning units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multi-view images are processed for 360-degree images, then the amount of data generated increases massively, but the performance of image processing systems becomes insufficient
Solution Approach 1:
The patent divides the 360-degree image into multiple view images captured by different cameras, and further segments the encoding process into separate encoding units for each view image. This segmentation allows the system to handle large amounts of data by processing smaller, manageable portions independently, thereby improving overall processing performance while maintaining comprehensive 360-degree coverage.
Solution Approach 2:
The patent transforms 2D view images into a 3D spherical representation through equirectangular projection and coordinate transformation. By adding the spatial dimension of viewing angle, the system can efficiently pack and organize image data in three-dimensional space, enabling better data management and processing performance despite the increased data quantity.
2Loss of energy
If 360-degree images are encoded and decoded, then compression performance is enhanced, but the complexity of the encoding and decoding process increases
Solution Approach 1:
The patent segments the 360-degree image into multiple view images and applies independent encoding units to each view. This segmentation simplifies the encoding process by allowing standard image encoding techniques to be applied to smaller, manageable portions, reducing the overall complexity while achieving compression through efficient data organization and packing.
Solution Approach 2:
The patent changes the parameter of image projection format to equirectangular projection and adjusts coordinate systems to match the spherical geometry. By standardizing these parameters, the system simplifies the encoding and decoding processes while maintaining compression efficiency, as the transformations become predictable and can be handled by standardized algorithms.
3Measurement precision
If image expansion is performed based on partitioning units, then the accuracy of predicted image generation is improved, but the processing time increases
Solution Approach 1:
The patent divides the image into partitioning units and performs image expansion independently on each unit. This segmentation allows the system to generate predicted images with high accuracy by processing each partition separately with appropriate expansion factors, while the overall processing time is managed through parallel processing of multiple partitions.
Solution Approach 2:
The patent applies image expansion selectively to regions that require it based on the partitioning units, rather than uniformly expanding the entire image. This partial action approach improves accuracy where needed while reducing unnecessary processing time in regions where expansion is not required, optimizing the balance between precision and efficiency.
Data Source
AI summary
Disclosed are methods and apparatuses for image data encoding/decoding. A method for decoding a 360-degree image includes the steps of: receiving a bitstream obtained by encoding a 360-degree image; generating a prediction image by making reference to syntax information obtained from the received bitstream; adding the generated prediction image to a residual image obtained by dequantizing and inverse-transforming the bitstream, so as to obtain a decoded image; and reconstructing the decoded image into a 360-degree image according to a projection format. Therefore, the performance of image data compression can be improved.