Adaptive and hierarchical visualization of segmented digital ink expressions based on context

Digital ink expressions can neither be converted into users' desired layouts nor be recognized in the semantic level until their symbols and hierarchical symbol relations are correctly extracted. There are many errors in automatically segmented digital ink expressions because they are free forms and mixed with various symbols, as well as the symbols have complex structures and hierarchical relations. Symbols and their relations in segmented digital ink expressions may be wrongly extracted. It is a prerequisite to visualize automatically segmented digital ink expressions before correcting them using human-computer interaction. An adaptive and hierarchical approach based on context is proposed to visualize automatically segmented digital ink expressions. Each extracted symbol is first adaptively visualized by color and shape labels according to overlap membership between it and its neighbors. The hierarchical relationship between symbols is visualized by a base tree with colored links. We have conducted experiments using real-life segmented digital ink expressions and compared the proposed approach with others. Experimental results demonstrate that the proposed approach is feasible, flexible, effective, and robust.