3D Pose Detection via 2D Contour Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in determining and configuring the appropriate device for consuming digital content, and existing technologies struggle to accurately detect and project content onto handheld display mediums, especially under conditions of partial occlusion.
Innovation Solution
The system employs augmented reality functional nodes (ARFNs) equipped with computing devices and interface components, including cameras and projectors, to detect the 3D pose of a handheld display medium by converting 3D point sets to 2D coordinates, aligning model contours with observed contours, and projecting content correctly, even under occlusion, using iterative closest point (ICP) minimization and keystone correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional device detection methods are used, then device identification is straightforward, but the system cannot accurately detect and track handheld display mediums in motion under partial occlusion
Solution Approach 1:
The system segments the display medium into multiple feature points distributed across its surface. By detecting and tracking these segmented feature points independently, the system can accurately determine the overall pose of the display medium even when portions are occluded, as long as sufficient feature points remain visible.
Solution Approach 2:
The system introduces an intermediary computational model that establishes geometric relationships between detected feature points and the underlying display medium parameters (position, orientation, scale). This intermediary model allows accurate pose estimation even when not all feature points are visible, resolving the contradiction between measurement precision and occlusion resistance.
2Productivity
If the system continuously tracks the display medium to enable seamless content projection, then user experience is improved, but computational complexity and processing requirements increase
Solution Approach 1:
The system performs preliminary action by pre-distributing multiple feature points across the display medium surface before tracking begins. This preliminary setup enables continuous tracking through simple geometric calculations rather than complex real-time analysis, maintaining productivity while managing device complexity.
Solution Approach 2:
The system creates a simplified computational copy of the display medium's geometry and pose parameters. Instead of directly analyzing complex visual data continuously, the system maintains and updates this simplified geometric model, enabling continuous tracking with reduced computational complexity.
3Reliability
If the system uses multiple feature points for robust pose detection, then detection reliability under occlusion is improved, but the complexity of contour alignment and processing increases
Solution Approach 1:
The system replaces complex mechanical-style contour alignment processes with direct geometric calculations based on feature point coordinates. By substituting the traditional contour-matching approach with point-based geometric relationships, the system achieves reliable pose detection through simpler computational operations.
Solution Approach 2:
The system changes the parameter representation from continuous contour descriptions to discrete feature point coordinates. This parameter transformation simplifies the alignment process by converting complex contour matching into straightforward geometric calculations involving point positions, distances, and angles, thereby improving reliability while reducing processing complexity.
Data Source
AI summary
Techniques are described for determining the pose of an object based on a 3D point set representing the object. The 3D point set is rotated into a 2D coordinate system and a model object contour is aligned with the contour of the rotated point set using an iterative process. The aligned model object contour is then rotated back into the original 3D coordinate system, where its pose is assumed to represent the pose of the object.