Point Cloud Encoding via Sensing Path Curve Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud compression technologies fail to effectively combine encoding and decoding simplicity, low latency, and compression performance, particularly for sparse geometry data sensed by spinning sensors, and do not adapt to the order of sensing, leading to suboptimal compression performance and increased latency.
Innovation Solution
The method encodes and decodes point clouds by representing the sensing path as a two-dimensional curve in a two-dimensional angular coordinate space, using control points for interpolation and coarse points to efficiently compress the data, allowing for flexible sampling and adaptation to regions of interest, thereby improving compression performance and reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing point cloud compression technologies are used, then encoding and decoding can be performed, but compression performance is suboptimal and latency is increased
Solution Approach 1:
The patent applies preliminary action by pre-defining a sensing path that represents the order in which points are sensed by the sensor. This sensing path is encoded and transmitted along with the point cloud data, allowing the decoder to process points in the correct sensing order without requiring complex analysis during decoding. This preliminary organization of data according to sensing sequence enables more efficient compression algorithms that exploit temporal and spatial correlations, thereby improving compression performance while reducing decoding latency.
Solution Approach 2:
The patent segments the point cloud data into groups based on the sensing path, organizing points according to their sensing order rather than treating them as a random set. This segmentation allows compression algorithms to process related points together, exploiting local correlations and reducing redundancy. By dividing the data into meaningful segments following the sensing trajectory, the system achieves better compression ratios and faster processing compared to traditional methods that don't account for sensing order.
2Ease of operation
If traditional encoding methods are used, then simplicity is maintained, but compression performance deteriorates
Solution Approach 1:
The encoding simplicity is maintained by pre-computing and storing the sensing path information during data acquisition. Rather than requiring complex real-time computations during encoding, the sensing path is established beforehand and simply referenced during the compression process. This preliminary action separates the complex path planning from the encoding operation, keeping the encoder itself simple while achieving improved compression performance through the organized data structure.
Solution Approach 2:
The sensing path acts as an intermediary structure that bridges the raw point cloud data and the compression algorithm. Instead of directly compressing unordered point data, the system introduces the sensing path as an intermediate representation that organizes points according to their acquisition sequence. This intermediary structure enables simple compression algorithms to achieve better performance by exploiting the ordered structure, without requiring complex encoding logic.
Data Source
AI summary
A method and apparatus is capable of encoding and decoding a point cloud sensed by any type of sensor following a sensing path. The method obtains coarse representations of sensed points and encodes controls points of a two-dimensional curve representing the sensing path and the coarse representations. The control points are decoded, the two-dimensional curve representing the sensing path is obtained by interpolation between the control points and coarse representations of points are decoded. Points of the point cloud are then reconstructed from the decoded two-dimensional curve representing the sensing path and the decoded coarse representations.