Automatic Museum Audio Guide

Area of Science:

• Computer Vision
• Human-Computer Interaction
• Museum Technology

Background:

• Traditional museum audio guides can be cumbersome and lack interactivity.
• There is a need for intuitive and efficient methods for visitors to access exhibit information.

Purpose of the Study:

• To develop and evaluate an automatic museum audio guide system.
• To assess the usability and user preference of the novel system compared to conventional audio guides.

Main Methods:

• The system utilizes a headset with a camera and an eyes of things (EoT) computer vision device for artwork recognition.
• Artwork recognition is achieved using accelerated segment test (FAST) keypoints and a random forest classifier.
• Application logic was implemented for efficient interaction upon artwork recognition, with testing in a real museum setting.

Main Results:

• The developed automatic audio guide system demonstrated effective artwork recognition.
• The system provided a full day of use on a single battery charge.
• User testing in a museum confirmed the system's simplicity and high user preference over traditional audio guides.

Conclusions:

• The automatic museum audio guide offers a user-friendly and efficient alternative to traditional systems.
• The eyes of things (EoT) device and associated algorithms provide a robust solution for real-time artwork identification.
• The system's design prioritizes ease of use, enhancing the overall museum visitor experience.