Point Cloud Visualization for Data Labeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual data labeling for machine learning models is time-consuming and labor-intensive, with few tools available to facilitate the process effectively.
Innovation Solution
A computer-implemented method for visualizing data that displays a 3D scene with colors determined by photographic, label, or LIDAR intensity data, allowing users to adjust blending of these data sources for improved labeling and verification of point clouds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual labeling is used, then labeling accuracy can be maintained, but time consumption and labor intensity increase significantly
Solution Approach 1:
The patent introduces an intermediary visualization system that combines multiple data sources (photographic data, LIDAR data, label data) to create a unified 3D point cloud representation. This intermediary layer allows users to verify and adjust labels more efficiently by seeing spatial relationships and contextual information that manual labeling would require to achieve the same accuracy.
Solution Approach 2:
The patent uses color changes in the point cloud visualization to represent different data sources and label states. By coloring points based on photographic data, LIDAR data, or label data, the system provides visual feedback that helps users verify labeling accuracy while working faster, as they can immediately see the impact of their labeling decisions in the 3D space.
2Reliability
If multiple data sources are combined for visualization, then labeling verification is improved, but system complexity increases
Solution Approach 1:
The patent segments the visualization system into distinct modules that handle different data sources (photographic data processing, LIDAR data processing, label data processing) separately, then combines them through a unified point cloud rendering engine. This segmentation allows each module to be optimized independently while maintaining overall system manageability, even as complexity increases.
Solution Approach 2:
The patent creates a universal visualization framework that can handle multiple types of data (photographic, LIDAR, label data) and multiple visualization modes (point cloud rendering, data source blending, verification modes) within a single system architecture. This multi-functionality reduces the need for separate specialized tools for each data type, thereby managing complexity while improving verification capabilities.
3Productivity
If point cloud colorings are blended from multiple data sources, then labeling efficiency improves, but processing requirements increase
Solution Approach 1:
The patent implements partial blending of point cloud colorings based on user needs and data availability. Rather than always processing all data sources at full resolution, the system can selectively blend only the necessary data sources (e.g., only photographic and LIDAR data when label verification is needed, or only label data when quick labeling is needed), reducing processing requirements while maintaining labeling efficiency for the given task.
Data Source
AI summary
One embodiment provides a user interface (UI) that permits users to select how point cloud colorings determined from multiple data sources are blended together in a rendering of a point cloud. The data sources may include photographic, label, and/or LIDAR intensity data. To improve frame rates, an aggregated point cloud may be generated using a spatial hash of a large set of points and sampling of each hash bucket based on the number of points therein and a user-configurable density. Sizes of points in the point cloud may decrease proportionally to distance from a viewer, but increase based on an activation function that enlarges points greater than a threshold distance from the viewer. In addition, luminance statistics for sub-regions of photographic data and dominant colors determined from photographic data may be used to automatically determine color properties to apply to a point cloud coloring.