Elevator Sign Language Translation With Camera-Based Generative AI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Elevator systems fail to effectively communicate with passengers who are mute and deaf during service outages, as conventional methods like emergency buttons or phones may not suffice, especially when the passenger is disabled.
Innovation Solution
Incorporating a camera, view screen, and communication device in the elevator car to observe and translate sign language movements into messages or sign language via a cloud-based AI model, enabling communication with operators and rescue personnel.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional emergency communication methods (buttons, phones) are used in elevator systems, then the system structure remains simple, but communication effectiveness with mute and deaf passengers is insufficient
Solution Approach 1:
The patent introduces an AI model as an intermediary between the camera (capturing sign language) and the communication system. The AI model translates sign language movements into text or speech, enabling effective communication with mute and deaf passengers without requiring the system to directly process complex sign language recognition
Solution Approach 2:
The patent replaces traditional mechanical communication interfaces (buttons, phones) with an AI-based visual recognition system. The camera captures sign language movements and the AI model processes these visual inputs, substituting the mechanical interaction paradigm with an intelligent visual processing paradigm that is more accessible to disabled passengers
2Adaptability or versatility
If sign language recognition and translation capabilities are added to communicate with disabled passengers, then communication accessibility is improved, but device complexity increases
Solution Approach 1:
The patent implements a multi-functional communication system that can handle multiple communication modes (text, speech, sign language recognition) through a single integrated AI model. This allows the system to adapt to different passenger needs (mute, deaf, or both) without requiring separate specialized systems for each disability type
Solution Approach 2:
The AI model serves as an intermediary that handles the complexity of sign language recognition and translation, shielding the rest of the system from this complexity. The camera captures raw video data and the AI model processes it into meaningful communication data, isolating the complexity management to a specific component
3Measurement precision
If cloud-based AI model is used to translate sign language movements, then translation accuracy is improved, but response time may be increased due to network communication
Solution Approach 1:
The patent pre-loads and caches the AI model in the local system before it is needed. When the elevator becomes trapped and sign language recognition is required, the system can immediately process video data without waiting for model loading or initialization, significantly reducing the perceived response time while maintaining cloud-based accuracy
Data Source
AI summary
An elevator system is provided. The elevator system includes an elevator car including a camera, a view screen and a communication device. In an event a passenger who is mute and deaf is trapped in the elevator car, communications via the communication device are initiated, sign language movements expressed by the passenger are observed by the camera whereupon the sign language movements are translated into messages for an operator and a sign language to which the sign language movements are associated is identified and responses from the operator are displayed to the passenger on the view screen.