Door-Opening Intent Detection Using Pose Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing door lock systems face challenges in determining user intent for opening, particularly with wireless interfaces like RFID, BLE, and UWB, which can lead to unauthorized access and inefficient energy use in automatic doors.
Innovation Solution
A method and system using image processing and machine learning to determine user intent by analyzing the position and orientation of a person through a camera, creating a stick figure representation, and using data structures to infer intent based on sequential image analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If RFID is used for wireless interface, then close proximity authentication is achieved, but large antenna size and high energy consumption are required
Solution Approach 1:
The patent replaces RFID's electromagnetic field-based authentication with a visual-based authentication system using camera imaging and pose estimation. Instead of relying on radio frequency signals and large antennas, the system uses optical imaging to detect and analyze human pose, thereby substituting a mechanical/optical system for the electromagnetic system and eliminating the need for large antennas.
2Adaptability or versatility
If longer range communication protocols like BLE or UWB are used, then remote system management is enabled, but intent determination becomes difficult leading to unauthorized access
Solution Approach 1:
The patent segments the authentication process into two distinct phases: proximity detection (determining if someone is near the door) and intent determination (determining if someone actually wants to open the door). This is achieved by analyzing sequential pose data to detect approach behavior followed by door-opening intent, thereby separating the functions of proximity sensing and intent recognition to improve access control reliability.
Solution Approach 2:
The system performs preliminary pose estimation and tracking before making the final authentication decision. By continuously monitoring and storing pose data points as a person approaches and interacts with the door, the system gathers preliminary information about behavior patterns, allowing it to reliably determine intent based on accumulated evidence rather than a single data point.
3Ease of operation
If automatic door opening is enabled without intent determination, then convenience is improved, but energy consumption increases due to unnecessary door openings
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors pose data and adjusts door opening decisions based on detected intent. The pose estimation system provides real-time feedback about user behavior, and only when specific intent patterns are detected (approach followed by door-opening gesture) does the system trigger door opening, thereby using feedback to eliminate unnecessary energy consumption while maintaining convenience.
Data Source
Figure 1~3
Figure 4
Figure 5~7
AI summary
It is provided a method for determining intent to open a door. The method is performed by an intent determiner. The method comprises: obtaining an image of a physical space near the door; determining a position and orientation of a person in the image by providing the image to an image machine learning model, wherein the image machine learning model is configured to determine position and orientation of a person in the image, wherein the image machine learning model is configured to determine a stick figure of the person based on the image; adding a data item, comprising an indicator of the position and an indicator of the orientation, to a data structure; determining based on the data structure, whether there is intent of the person to open the door; and repeating the method until an exit condition is true.