Saliency-Assisted Navigation of Very Large Landscape Images

Abstract

Abstract-The field of visualization has addressed navigation of very large datasets, usually meshes and volumes. Significantly less attention has been devoted to the issues surrounding navigation of very large images. In the last few years the explosive growth in the resolution of camera sensors and robotic image acquisition techniques has widened the gap between the display and image resolutions to three orders of magnitude or more. This paper presents the first steps towards navigation of very large images, particularly landscape images, from an interactive visualization perspective. The grand challenge in navigation of very large images is identifying regions of potential interest. In this paper we outline a three-step approach. In the first step we use multi-scale saliency to narrow down the potential areas of interest. In the second step we outline a method based on statistical signatures to further cull out regions of high conformity. In the final step we allow a user to interactively identify the exceptional regions of high interest that merit further attention. We show that our approach of progressive elicitation is fast and allows rapid identification of regions of interest. Unlike previous work in this area, our approach is scalable and computationally reasonable on very large images. We validate the results of our approach by comparing them to user-tagged regions of interest on several very large landscape images from the Internet.

Guided Visual Analytics for Image Selection in Time and Space

“…directing -providing a ranked set of suggestions, common in applications where different options can be clearly defined (e.g., sampling from a large parameter space, recommending different actions and analysis paths [11]); and prescribing, or automated and self-enacted decision-making, more commonly used in analytical systems to provide a starting point or path for analysis (e.g., touring the user through calculated interest points in very large images [15]). These types of guidance were recently abstracted into system guidance tasks, coupled to the analysis of user tasks, and further expanded into a typology by Pérez-Messina et al [23], on which we base our task analysis (Sec.…”

A Typology of Guidance Tasks in Mixed‐Initiative Visual Analytics Environments

“…An observation can be anything that stems from a user decision, e.g., a change in the current view, a movement of the mouse, and also the lack of interaction. However, a guidance system can also provide guidance without considering user-input, i.e., without making adaptations to the guidance (e.g., in Ip and Varshney [IV11]). These latter systems provide guidance based mostly on the dataset under analysis, and make their suggestions available from the beginning of the session, suggestions which will not undergo any changes unless the data under analysis does.…”

“…Prescribing guidance is the degree with the highest chance of incurring in information loss, i.e., of nullifying possible analytic paths stemming from the user. We can find prescribing guidance in Horvitz et al [HBH * 13] and Ip and Varshney [IV11].…”

Challenges and opportunities: from big data to knowledge in AI 2.0

“…Interaction techniques such as focus + context (Baudisch et al, 2002), overview + details (Shneiderman, 1996), and saliency-aware navigation (Kim and Varshney, 2006) were proposed to provide multiple perspectives of data patterns according to the users' demand. In the exploration of large-scale landscape images (gigapixels), saliency-guided navigation (Ip and Varshney, 2011) seems to be a reasonable solution to rapid identification of regions of interest. Although users are able to express their data requirements with interactive visualization, they are not able to contribute their domain knowledge to visualization for better data understanding.…”