3D Spatial Region and Atlas Tile Group Linking in Point Cloud Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for video-based point cloud compression (V-PCC) lack a defined signaling mechanism for linking 3D spatial regions with atlas tile groups, making it difficult to decode and reconstruct partial point clouds efficiently.
Innovation Solution
An image processing apparatus and method that selects and links 3D spatial regions with corresponding atlas tile groups using stored link information, extracts and decodes bitstreams and NAL units to generate 2D data, and encodes 2D data to generate bitstreams with link information for efficient decoding and reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If V-PCC bitstream is stored in ISOBMFF without link information, then file compatibility is improved, but the ability to link 3D spatial region with atlas tile group deteriorates
Solution Approach 1:
The patent segments the ISOBMFF file structure into distinct boxes: a first box storing the V-PCC bitstream and a second box storing link information that associates 3D spatial regions with atlas tile groups. This segmentation allows the file to maintain compatibility with ISOBMFF standards while incorporating the necessary linking information without disrupting the overall file structure.
Solution Approach 2:
The patent introduces link information as an intermediary element that mediates between the V-PCC bitstream and the atlas tile groups. This link information acts as a bridge, enabling the decoder to correctly associate 3D spatial regions with their corresponding atlas tile groups without modifying the core V-PCC encoding or the ISOBMFF structure itself.
2Adaptability or versatility
If link information is added to ISOBMFF file, then linking capability is improved, but device complexity deteriorates
Solution Approach 1:
By segmenting the file structure into distinct functional boxes (V-PCC bitstream box and link information box), the patent adds linking capability without creating a monolithic complex structure. Each box can be processed independently, and the segmentation follows ISOBMFF conventions, making the added complexity manageable and structured.
Solution Approach 2:
The link information is prepared and organized in advance during the encoding stage, storing the associations between 3D spatial regions and atlas tile groups in a predetermined format. This preliminary organization reduces the complexity burden on the decoding device, as the linking relationships are already established and do not require complex runtime computation.
3Productivity
If partial point cloud decoding is implemented without link information, then decoding speed is improved, but reconstruction accuracy deteriorates
Solution Approach 1:
The patent extracts and stores the essential linking relationships between 3D spatial regions and atlas tile groups in a separate link information box. This extraction allows the decoder to quickly access only the necessary linking data for partial point cloud decoding, maintaining decoding speed while ensuring that the correct atlas tile groups are selected for accurate reconstruction of the specified 3D spatial region.
Solution Approach 2:
The associations between 3D spatial regions and atlas tile groups are predetermined and stored in the link information during encoding. This preliminary setup enables the decoder to rapidly retrieve the correct mappings without performing complex computations during decoding, thus maintaining high decoding speed while ensuring reconstruction accuracy through proper region-tile group correspondence.
Data Source
AI summary
The present disclosure relates to an image processing apparatus and method that enable linking between a 3D spatial region and an atlas tile group. Link information is generated and signaled, in which the link information is stored in a file together with a bitstream of a point cloud expressing an object having a three-dimensional shape as a set of points, and the link information links a 3D spatial region of the point cloud with an atlas tile group. On the basis of the link information, the 3D spatial region and the atlas tile group that correspond to a desired three-dimensional space are selected. The present disclosure can be applied to, for example, an image processing apparatus, an information processing apparatus, an image processing method, an information processing method, or the like.