Abstract

Documenting underwater cultural heritage is a challenging undertaking. Underwater environment is not a man's natural habitat and special equipment and devices had to be invented so that he could enter and study this environment. Several decades of underwater research and many sacrifices were needed to fully understand the importance of underwater heritage and its protection. The means for accurate documentation underwater are very limited and demanding, due to required technical equipment it is also expensive. Emergence of modern 3D methods and accompanying software tools for processing of 3D data is therefore of utmost importance for documenting and protection of underwater cultural heritage. In comparison to manual and analog methods, 3D methods offer much better accuracy, they substantially shorten the necessary time spent underwater and in this way improve the safety at work as well as lower the entire cost of field work. For illustration of the above development we discuss archeological case studies from the North East Adriatic.

The paper presents an approach to using structural descriptions, obtained through a human-robot tutoring dialogue, as labels for the visual object models a robot learns. The paper shows how structural descriptions enable relating models for different aspects of one and the same object, and how being able to relate descriptions for visual models and discourse referents enables incremental updating of model descriptions through dialogue (either robot-or human-initiated). The approach has been implemented in an integrated architecture for human-assisted robot visual learning.

No abstract

No abstract