Depth Map Confidence Scoring for Planar-Region 3D Meshes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating 3D meshes face inaccuracies due to planar regions and fast device movements, leading to faulty artifacts and inaccurate depth measurements, particularly in indoor environments.
Innovation Solution
A system that generates confidence scores for depth estimation values, using factors like depth consistency, color variation, and truncation distances to identify and correct unreliable depth measurements, thereby improving the accuracy of 3D mesh generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If depth measurements are obtained in indoor environments with planar regions, then 3D mesh generation can be performed, but measurement precision deteriorates due to planar regions and fast device movements causing faulty artifacts
Solution Approach 1:
The system performs preliminary actions by obtaining multiple depth measurements before final mesh generation, including initial depth map acquisition, artifact detection, and preliminary filtering. This allows the system to identify and correct potential measurement errors before they propagate to the final 3D mesh, thereby improving both precision and reliability.
Solution Approach 2:
The system implements feedback mechanisms by detecting artifacts in depth measurements and using this information to adjust subsequent measurements. The artifact detection results feed back into the measurement process, allowing the system to refine depth estimates and improve accuracy in challenging indoor environments with planar regions.
2Measurement precision
If multiple depth measurements are obtained to improve accuracy, then measurement precision improves, but loss of time increases due to multiple measurement cycles
Solution Approach 1:
The system applies partial action by performing depth measurements selectively rather than uniformly across all areas. It focuses measurement resources on regions where artifacts are detected or where measurement uncertainty is high, while reducing measurements in already well-characterized areas. This approach maintains accuracy where needed while minimizing overall measurement time.
Solution Approach 2:
The system segments the depth measurement process into distinct phases: initial rapid scanning, artifact detection, targeted re-measurement of problematic regions, and final mesh generation. This segmentation allows the system to spend more time only on critical measurement areas rather than uniformly increasing measurement time across the entire scene.
Data Source
AI summary
Embodiments of devices and techniques of obtaining a three dimensional (3D) representation of an area are disclosed. In one embodiment, a two dimensional (2D) frame is obtained of an array of pixels of the area. Also, a depth frame of the area is obtained. The depth frame includes an array of depth estimation values. Each of the depth estimation values in the array of depth estimation values corresponds to one or more corresponding pixels in the array of pixels. Furthermore, an array of confidence scores is generated. Each confidence score in the array of confidence scores corresponds to one or more corresponding depth estimation values in the array of depth estimation values. Each of the confidence scores in the array of confidence scores indicates a confidence level that the one or more corresponding depth estimation values in the array of depth estimation values is accurate.