Interactive unsupervised classification and visualization for browsing an image collection

Bruneau, Pierrick; Picarougne, Fabien; Gelgon, Marc

doi:10.1016/j.patcog.2009.03.024

Cited by 8 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gomi and Itoh [6] use a 3D visualization to present image retrieval results on a mobile device, where thumbnails are arranged along the Z axis according to their context-dependent priority. A similar idea was proposed by Bruneau et al [3] for visualization of a cluster graph of images. The Photo Explorer proposed by Snavely et al [13] arranges photos in a 3D layout according to geographic coordinates.…”

Section: Related Workmentioning

confidence: 97%

The Stack-of-Rings Interface for Large-Scale Image Browsing on Mobile Touch Devices

Schoeffmann

2014

Proceedings of the 22nd ACM International Conference on Multimedia

View full text Add to dashboard Cite

We propose a new interface for browsing large image collections on touch-enabled devices, which allows for faster visual search than a common grid interface. The proposed interface consists of 3D rings arranged in a vertical stack. Each ring can be rotated independently with a horizontal wipe gesture and the whole stack can be moved up and down with vertical wipe gestures. We present results from a user study with 16 participants, where we compared the performance of the proposed interface to the one provided by the default image browsing interface on tablets. The evaluation results show that the proposed interface performs significantly faster than the default interface on the tablet and achieves a significantly better rating by the users.

show abstract

Section: Related Workmentioning

confidence: 97%

The Stack-of-Rings Interface for Large-Scale Image Browsing on Mobile Touch Devices

Schoeffmann

2014

Proceedings of the 22nd ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…This local optimization can be interpreted as the temporal evolution of a 2D graph completely linked by a system of springs. This relates t-SNE to spring-based graph layout algorithms [37], already employed as a DR method in the literature [22,13,4]. Clustering usually takes place in the HD space, and its results are then overlaid on a 2D projection for off-line inspection.…”

Section: A Summary Of the T-sne Projection Techniquementioning

confidence: 99%

“…Such visual representations can be used both for building and adjusting a mental map of the data at hand, or as an entry point for finer data inspection using brushing techniques [12]. This need may occur when considering the organization of image collections [13], in the context of multimedia retrieval 3 engines, or when reporting public or medical data. In many cases though, the initial clustering of a data set is deemed to be imperfect with regard to a ground truth, or user expectations.…”

mentioning

confidence: 99%

Cluster Sculptor, an interactive visual clustering system

Bruneau¹,

Pinheiro²,

Broeksema³

et al. 2015

Neurocomputing

View full text Add to dashboard Cite

International audienceThis paper describes Cluster Sculptor, a novel interactive clustering system that allows a user to iteratively update the cluster labels of a data set, and an as-sociated low-dimensional projection. The system is fed by clustering results computed in a high-dimensional space, and uses a 2D projection, both as sup-port for overlaying the cluster labels, and engaging user interaction. By easily interacting with elements directly in the visualization, the user can inject his or her domain knowledge progressively, crafting an updated 2D projection and the associated clustering structure that combine his or her preferences and the manifolds underlying the data. Via interactive controls, the distribution of the data in the 2D space can be used to amend the cluster labels, or reciprocally, the 2D projection can be updated so as to emphasize the current clusters. The 2D projection updates follow a smooth physical metaphor, that gives insight of the process to the user. Updates can be interrupted any time, for further data inspection, or modifying the input preferences. The interest of the system is demonstrated by detailed experimental scenarios on three real data sets

show abstract

“…However, for shared image repositories where multiple people contribute images and use the photographs it can be a huge challenge to find exactly the images one is looking for. The literature focuses on three approaches to handling large image collections, namely manual image tagging [1,2], content based image retrieval [3][4][5] and image browsing [6][7][8]. Manual image tagging is the strategy whereupon an image or a collection is manually tagged with keywords or text.…”

Section: Introductionmentioning

confidence: 99%

“…Browsing applications such as Picasa has also employed techniques from image processing, in particular face recognition, to assist in the image browsing process. The merging of ideas from browsing and image processing is an emerging trend [6][7][8] and this is also the basis for the approach taken in this paper.…”

Section: Introductionmentioning

confidence: 99%

A Configurable Photo Browser Framework for Large Image Collections

Sandnes

2011

Human-Computer Interaction. Design and Development Approaches

View full text Add to dashboard Cite

Abstract. Image collections are growing at an exponential rate due to the wide availability of inexpensive digital cameras and storage. Current browsers organize photos mostly chronologically, or according to manual tags. For very large collections acquired over several years it can be difficult to locate a particular set of images -even for the owner. Although our visual memory is powerful, it is not always easy to recall all of one's images. Moreover, it can be very time consuming to find particular images in other peoples image collections. This paper presents a prototype image browser and a plug-in pattern that allows classifiers to be implemented and easily integrated with the image browser such that the user can control the characteristics of the images that are browsed and irrelevant photos are filtered out. The filters can both be content based and based on meta-information. The current version is only employs meta-information which means that large image collections can be indexed efficiently.

show abstract

Interactive unsupervised classification and visualization for browsing an image collection

Cited by 8 publications

References 19 publications

The Stack-of-Rings Interface for Large-Scale Image Browsing on Mobile Touch Devices

The Stack-of-Rings Interface for Large-Scale Image Browsing on Mobile Touch Devices

Cluster Sculptor, an interactive visual clustering system

A Configurable Photo Browser Framework for Large Image Collections

Contact Info

Product

Resources

About