Building and tracking hierarchical geographical &amp; temporal partitions for image collection management on mobile devices

Pigeau, Antoine; Gelgon, Marc

doi:10.1145/1101149.1101170

Cited by 36 publications

(42 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a third pass, temporal segments that belong to the same location cluster are merged. Pigeau, et al [3,10], also present a multi-pass system. The first pass performs clustering using mixture models learned jointly on the time and location data with a variational approach to determine model order.…”

Section: Related Workmentioning

confidence: 99%

Clustering geo-tagged photo collections using dynamic programming

Cooper

2011

Proceedings of the 19th ACM International Conference on Multimedia

View full text Add to dashboard Cite

This paper describes methods for clustering photos that possess both time stamps and geographical coordinates as metadata. We present a two part method that first analyzes photos' time and location information to independently partition the photos into multiple clusterings. A subset of the detected clusters is then selected for the final photo clustering using an efficient dynamic programming procedure that optimizes a clustering fitness score. We propose fitness measures to produce clusterings that are coherent in space, time, or both. One group of scores directly measures within-cluster inter-photo distances. A second set of scores measures clusters' consistency with the reference clusterings. We present experiments that validate our method using multiple data sets.

show abstract

Section: Related Workmentioning

confidence: 99%

Clustering geo-tagged photo collections using dynamic programming

Cooper

2011

Proceedings of the 19th ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…The RED algorithm allows photos to be sorted, creating collections based on time and location. The RED algorithm follows on a large amount of related work that attempts to organise photo collections based on a variety of variables, including time and location [18,19,20,21]. Most existing algorithms are unsuitable in a mobile environment, as they are resource intensive and may result in significant delays before photos are sorted.…”

Section: Information Adaptationmentioning

confidence: 99%

Design and Evaluation of an Adaptive Mobile Map-Based Visualisation System

Tonder

Wesson

2009

Human-Computer Interaction – INTERACT 2009

View full text Add to dashboard Cite

Abstract. Mobile visualisation of map-based information is a difficult task. Designers of such systems must contend with the limitations of mobile devices in terms of hardware, screen size and input mechanisms. These problems are exacerbated by the nature of spatial data, where a large information space needs to be presented and manipulated on a small screen. In this paper, a prototype adaptive mobile map-based visualisation system, called MediaMaps, is presented. MediaMaps allows users to capture, location-tag, sort and browse multimedia in a map-based view. MediaMaps was designed to adapt the information visualised, the map-based visualisations and the supporting user interface. The results of an international field study, in which participants used MediaMaps on their personal mobile phones for a three-week period, are also presented. These results show that the adaptations implemented achieved high levels of accuracy and user satisfaction and successfully addressed some of the limitations of mobile map-based visualisation.

show abstract

“…The main steps of our approach are: 1.incremental building of a hierarchical temporal classification: we propose an improvement of our incremental and hierarchical algorithm [1], which provides a classification from the time stamp of each image. With our approach, we tend to emphasize the browsing task, rather than querying (a point motivated by the partial memory that the user has of the collection).…”

Section: Goal and Existing Workmentioning

confidence: 99%