2017 IEEE Pacific Visualization Symposium (PacificVis) (2017)
Seoul, South Korea
April 18, 2017 to April 21, 2017
Benjamin Niedermann , Karlsruhe Institute of Technology, Germany
Martin Nollenburg , TU Wien, Austria
Ignaz Rutter , TU Eindhoven, Netherlands
The usefulness of technical drawings as well as scientific illustrations such as medical drawings of human anatomy essentially depends on the placement of labels that describe all relevant parts of the figure. In order to not spoil or clutter the figure with text, the labels are often placed around the figure and are associated by thin connecting lines to their features, respectively. This labeling technique is known as external label placement. In this paper we introduce a flexible and general approach for external label placement assuming a contour of the figure prescribing the possible positions of the labels. While much research on external label placement aims for fast labeling procedures for interactive systems, we focus on highest-quality illustrations. Based on interviews with domain experts and a semi-automatic analysis of 202 handmade anatomical drawings, we identify a set of 18 layout quality criteria, naturally not all of equal importance. We design a new geometric label placement algorithm that is based only on the most important criteria. Yet, other criteria can flexibly be included in the algorithm, either as hard constraints not to be violated or as soft constraints whose violation is penalized by a general cost function. We formally prove that our approach yields labelings that satisfy all hard constraints and have minimum overall cost. Introducing several speedup techniques, we further demonstrate how to deploy our approach in practice. In an experimental evaluation on real-world anatomical drawings we show that the resulting labelings are of high quality and can be produced in adequate time.
Labeling, Layout, Human anatomy, Mathematical model, Interviews, Cost function
B. Niedermann, M. Nollenburg and I. Rutter, "Radial contour labeling with straight leaders," 2017 IEEE Pacific Visualization Symposium (PacificVis)(PACIFICVIS), Seoul, South Korea, 2017, pp. 295-304.