Systems and methods for automated design of camera placement and cameras arrangements for autonomous checkout
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for tracking subjects and detecting puts and takes of items in real space, such as cashier-less shopping systems, face challenges in determining optimal camera placement to ensure effective coverage while minimizing costs, particularly in large spaces with multiple moving subjects and occlusions.
Innovation Solution
A computer-implemented method that uses machine learning to iteratively optimize the number and pose of cameras in a three-dimensional space, applying objective functions and constraints to improve camera coverage scores, allowing for reduced or unchanged camera numbers while maintaining effective tracking of subjects and interactions with items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of cameras is increased to improve coverage of the area, then the coverage score is improved, but the installation and operation cost increases
Solution Approach 1:
The system changes parameters such as camera pose, position, and orientation to optimize coverage. By adjusting these parameters iteratively through machine learning, the system achieves improved coverage scores without necessarily increasing the number of cameras, thus resolving the contradiction between coverage quality and quantity of cameras.
Solution Approach 2:
The camera placement plan is made dynamic and adaptive through iterative machine learning optimization. The system can adjust camera poses and positions based on changing requirements and performance metrics, allowing optimal coverage to be achieved with minimized camera quantity rather than relying on static over-provisioning.
2Reliability
If manual optimization of camera placement is performed, then coverage can be improved, but time and computational resources are consumed
Solution Approach 1:
The system performs self-optimization of camera placement using automated machine learning processes. The iterative optimization is conducted autonomously without requiring continuous manual intervention, allowing the system to improve its own camera placement configuration while minimizing the time and resources required for optimization.
3Reliability
If cameras are placed to cover occluded areas, then detection reliability is improved, but the number of cameras required increases
Solution Approach 1:
The system applies local quality optimization by focusing camera placement and orientation on specific areas that require coverage, particularly occluded regions. Rather than uniformly distributing cameras throughout the space, the machine learning process identifies and targets specific locations and angles that maximize detection reliability in problematic areas, achieving improved reliability without proportionally increasing the total number of cameras.
Data Source
AI summary
Systems and techniques are provided for determining an improved camera coverage plan including a number, a placement, and a pose of cameras that are arranged to track puts and takes of items by subjects in a three-dimensional real space. The method includes receiving an initial camera coverage plan including a three-dimensional map of a real space, an initial number and initial pose of a plurality of cameras and a camera model including characteristics of the cameras. The method can iteratively apply a machine learning process to an objective function of number and poses of cameras, and subject to a set of constraints, obtain an improved camera coverage plan. The improved camera coverage plan is provided to an installer to arrange cameras to track puts and takes of items by subjects in the three-dimensional real space.