Temporal Point-Cloud Attribute Coding to Reduce Data Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods for point clouds with multiple attributes fixedly assigned across frames lead to increased data coding and processing loads, despite not all attributes being necessary for every frame.
Innovation Solution
The number of attributes corresponding to a single geometry in a point cloud is made variable in the time direction by selectively coding only necessary attribute video frames, using attribute information to signal which frames are coded.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple attributes are fixedly assigned to all frames, then complete attribute information is preserved, but data volume and processing load increase
Solution Approach 1:
The patent extracts and codes only the necessary attribute video frames rather than all attribute frames. By identifying and selecting only those attributes that are actually needed for the current frame, the system removes redundant attribute data, thereby reducing data volume while preserving essential attribute information.
Solution Approach 2:
Instead of processing all attributes for every frame (excessive action), the patent applies partial action by selectively coding only the necessary attributes for each frame. This approach processes fewer items than the maximum possible, optimizing the balance between information completeness and data efficiency.
2Loss of information
If all attribute frames are coded, then no attribute information is lost, but coding load and transmission bandwidth increase
Solution Approach 1:
The system extracts and identifies necessary attributes for each frame, then codes only those specific attributes rather than all attributes. This extraction approach prevents information loss for necessary attributes while eliminating redundant coding operations, thereby improving coding efficiency.
Solution Approach 2:
The patent introduces dynamic attribute selection where the set of coded attributes changes based on frame-specific requirements. Rather than a static all-or-nothing approach, the system dynamically adjusts which attributes are coded for each frame, optimizing the balance between information preservation and coding efficiency.
3Manufacturing precision
If the number of attributes is increased for all frames, then attribute detail and quality improve, but processing complexity and data size increase
Solution Approach 1:
The patent applies local quality by allowing different numbers of attributes to be assigned to different frames based on their specific needs. Rather than uniformly increasing attributes across all frames, the system selectively enhances attribute detail only where necessary, thereby improving quality without proportionally increasing processing complexity.
Data Source
AI summary
There is provided an image processing device and a method that enables the number of attributes corresponding to a single geometry to be variable in a time direction. An attribute video frame that is a video frame in which a patch obtained by projecting each of a plurality of attributes to a two-dimensional plane for each partial region is arranged is generated, the plurality of attributes corresponding to a single geometry of a point cloud that expresses an object with a three-dimensional shape as a group of points, the generated attribute video frame for each attribute is coded, and attribute information that is information indicating the generated attribute video frames corresponding to the same timing is generated. The present disclosure can be applied to, for example, an image processing device, electronic equipment, an image processing method, a program, or the like.