IEEE Trans Vis Comput Graph. 2020 Oct;26(10):3063-3076. doi: 10.1109/TVCG.2019.2908166. Epub 2019 Mar 29.

VeCHArt: Visually Enhanced Comparison of Historic Art Using an Automated Line-Based Synchronization Technique.

IEEE transactions on visualization and computer graphics

Hermann Pfluger, Dennis Thom, Anna Schutz, Daniela Bohde, Thomas Ertl

PMID: 30946669 DOI: 10.1109/TVCG.2019.2908166

Abstract

The analysis of subtle deviations between different versions of historical prints has been a long-standing challenge in art history research. So far, this challenge has required extensive domain knowledge, fine-tuned expert perception, and time-consuming manual labor. In this paper we introduce an explorative visual approach to facilitate fast and accurate support for the task of comparing differences between prints such as engravings and woodcuts. To this end, we have developed a customized algorithm that detects similar stroke-patterns in prints and matches them in order to allow visual alignment and automated deviation highlighting. Our visual analytics system enables art history researchers to quickly detect, document, and categorize qualitative and quantitative discrepancies, and to analyze these discrepancies using comprehensive interactions. To evaluate our approach, we conducted a user study involving both experts on historical prints and laypeople. Using our new interactive technique, our subjects found about 20 percent more differences compared to regular image viewing software as well as "paper-based" comparison. Moreover, the laypeople found the same differences as the experts when they used our system, which was not the case for conventional methods. Informal feedback showed that both laypeople and experts strongly preferred employing our system to working with conventional methods.

Publication Types

• Journal Article