Demonstration venue · read-only. Every page can be browsed; the buttons that would change it are switched off. Create an account to run TaxoReview on your own data. A simulation from public publication records: the people listed were not invited and did not agree to review.
? Tutorial for this page

Lily Shirvanee

dblp:17/4392 · DBLP profile ↗
← Back
1ranked-venue papers
1as first author
0since 2021 · last 2003
—ORCID · none

Domains — the database's venue-derived domains; a paper can count in several

Graphics, computer vision, multimedia, augmented reality and games · 1 · 1 first-authorHuman-computer interaction and ubiquitous computing · 1 · 1 first-author

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Human-computer interaction and pervasive computing
1 paper
Ubiquitous computing and smart environments · 44% Interaction techniques and input · 44% Wearable and physiological sensing · 13%

Topics — the 3 heaviest of 3, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
Interaction techniques and input
gesture input
0.012003
The ViscousDisplay: adaptive transient interfaces in public space · SIGGRAPH 2003
Ubiquitous computing and smart environments › public displays
public interactive display
0.012003
The ViscousDisplay: adaptive transient interfaces in public space · SIGGRAPH 2003
Wearable and physiological sensing
multimodal sensing
0.012003
The ViscousDisplay: adaptive transient interfaces in public space · SIGGRAPH 2003

Methods — techniques the papers use, named apart from their topics

learning algorithms · 0.0
YearPublicationVenuePosition
2003 The ViscousDisplay: adaptive transient interfaces in public space
abstract
The ViscousDisplay explores the exchange of social information through transient public interfaces. Shaped by principles of 'underground public art', the ViscousDisplay is conceived as a novel mobile communication medium, where messages can be shared in public spaces. Inspired by biological learning systems; the ViscousDisplay learns gestural motions and colors that form along traces of a participant's movements and maps this information onto a flexible display. Because it is made up of inexpensive materials, the ViscousDisplay is also a disposable artifact which may be collected in public spaces. It combines multi-modal sensing, learning algorithms, and a pliable silicone display.
Lily Shirvanee, Marc Downie
SIGGRAPH1