Exploring Architectural Details Through a Wearable Egocentric Vision Device.

Area of Science:

• Computer Vision
• Augmented Reality
• Cultural Heritage

Background:

• Growing demand for augmented user experiences in cultural heritage.
• Need for enhanced visitor engagement with digital native tourists.
• Limitations of current methods in providing real-time, context-aware information.

Purpose of the Study:

• To propose a novel solution for augmenting outdoor cultural heritage tours.
• To assist visitors by providing location-specific information using wearable technology.
• To leverage computer vision for detailed landmark analysis and user guidance.

Main Methods:

• Utilizing computer vision techniques with a robust descriptor based on local feature covariance.
• Employing a lightweight wearable board for user localization against a 3D point cloud of the landmark.
• Developing a system to retrieve and present information about details the user is viewing.

Main Results:

• Experimental validation demonstrating the method's accuracy and computational efficiency.
• Successful localization of users within the 3D point cloud of historical sites.
• Accurate retrieval and presentation of relevant details to the user.

Conclusions:

• The proposed system effectively enriches the cultural heritage experience for visitors.
• Wearable cameras combined with computer vision offer a viable solution for augmented tourism.
• The method is accurate, computationally efficient, and positively evaluated by users.