Zoomless Maps: External Labeling Methods for the Interactive Exploration of Dense Point Sets at a Fixed Map Scale

TL;DR

This paper introduces external labeling methods for dense point set visualization on small screens, enabling navigation without zooming by placing labels at the map boundary and providing interaction techniques to access all features.

Contribution

It presents a unified model and algorithmic framework for external labeling with interactive methods, including exact and heuristic algorithms, tailored for dense spatial data visualization on small devices.

Findings

Algorithms effectively optimize label placement based on various criteria.

Interaction techniques improve feature accessibility without zooming.

Framework is flexible for different labeling variants.

Abstract

Visualizing spatial data on small-screen devices such as smartphones and smartwatches poses new challenges in computational cartography. The current interfaces for map exploration require their users to zoom in and out frequently. Indeed, zooming and panning are tools suitable for choosing the map extent corresponding to an area of interest. They are not as suitable, however, for resolving the graphical clutter caused by a high feature density since zooming in to a large map scale leads to a loss of context. Therefore we present new external labeling methods that allow navigating through dense sets of points of interest while keeping the current map extent fixed. We provide a unified model, in which labels are placed at the boundary of the map and visually associated with the corresponding features via connecting lines, which are called leaders. Since the screen space is limited, labeling all features at the same time is impractical. Therefore, at any time, we label a subset of the features. We offer interaction techniques to change the current selection of features systematically and, thus, give the user access to all features. We distinguish three methods, which allow the user either to slide the labels along the bottom side of the map or to browse the labels based on pages or stacks. We present a generic algorithmic framework that provides us with the possibility of expressing the different variants of interaction techniques as optimization problems in a unified way. We propose both exact algorithms and fast and simple heuristics that solve the optimization problems taking into account different criteria such as the ranking of the labels, the total leader length and the distance between leaders. We experimentally evaluate these algorithms and discuss the three variants with respect to their strengths and weaknesses proving the flexibility of the presented algorithmic framework.