Intelligent Temporal Subsampling of American Sign Language Using Event Boundaries

David H. Parish, George Sperling, Michael Landy

Research output: Contribution to journalArticle

Abstract

How well can a sequence of frames be represented by a subset of the frames? Video sequences of American Sign Language (ASL) were investigated in two modes: dynamic (ordinary video) and static (frames printed side by side on the display). An activity index was used to choose critical frames at event boundaries, times when the difference between successive frames is at a local minimum. Sign intelligibility was measured for 32 experienced ASL signers who viewed individual signs. For full gray-scale dynamic signs activity-index subsampling yielded sequences that were significantly more intelligible than when every mth frame was chosen. This result was even more pronounced for static images. For binary images, the relative advantage of activity subsampling was smaller. We conclude that event boundaries can be defined computationally and that subsampling from event boundaries is better than choosing at regular intervals.

Original languageEnglish (US)
Pages (from-to)282-294
Number of pages13
JournalJournal of Experimental Psychology: Human Perception and Performance
Volume16
Issue number2
StatePublished - May 1990

Fingerprint

Sign Language
American Sign Language
Subsampling

ASJC Scopus subject areas

  • Experimental and Cognitive Psychology
  • Behavioral Neuroscience
  • Cognitive Neuroscience

Cite this

Intelligent Temporal Subsampling of American Sign Language Using Event Boundaries. / Parish, David H.; Sperling, George; Landy, Michael.

In: Journal of Experimental Psychology: Human Perception and Performance, Vol. 16, No. 2, 05.1990, p. 282-294.

Research output: Contribution to journalArticle

@article{fd0124155cdb40949a9de30246ae3e3e,
title = "Intelligent Temporal Subsampling of American Sign Language Using Event Boundaries",
abstract = "How well can a sequence of frames be represented by a subset of the frames? Video sequences of American Sign Language (ASL) were investigated in two modes: dynamic (ordinary video) and static (frames printed side by side on the display). An activity index was used to choose critical frames at event boundaries, times when the difference between successive frames is at a local minimum. Sign intelligibility was measured for 32 experienced ASL signers who viewed individual signs. For full gray-scale dynamic signs activity-index subsampling yielded sequences that were significantly more intelligible than when every mth frame was chosen. This result was even more pronounced for static images. For binary images, the relative advantage of activity subsampling was smaller. We conclude that event boundaries can be defined computationally and that subsampling from event boundaries is better than choosing at regular intervals.",
author = "Parish, {David H.} and George Sperling and Michael Landy",
year = "1990",
month = "5",
language = "English (US)",
volume = "16",
pages = "282--294",
journal = "Journal of Experimental Psychology: Human Perception and Performance",
issn = "0096-1523",
publisher = "American Psychological Association Inc.",
number = "2",

}

TY - JOUR

T1 - Intelligent Temporal Subsampling of American Sign Language Using Event Boundaries

AU - Parish, David H.

AU - Sperling, George

AU - Landy, Michael

PY - 1990/5

Y1 - 1990/5

N2 - How well can a sequence of frames be represented by a subset of the frames? Video sequences of American Sign Language (ASL) were investigated in two modes: dynamic (ordinary video) and static (frames printed side by side on the display). An activity index was used to choose critical frames at event boundaries, times when the difference between successive frames is at a local minimum. Sign intelligibility was measured for 32 experienced ASL signers who viewed individual signs. For full gray-scale dynamic signs activity-index subsampling yielded sequences that were significantly more intelligible than when every mth frame was chosen. This result was even more pronounced for static images. For binary images, the relative advantage of activity subsampling was smaller. We conclude that event boundaries can be defined computationally and that subsampling from event boundaries is better than choosing at regular intervals.

AB - How well can a sequence of frames be represented by a subset of the frames? Video sequences of American Sign Language (ASL) were investigated in two modes: dynamic (ordinary video) and static (frames printed side by side on the display). An activity index was used to choose critical frames at event boundaries, times when the difference between successive frames is at a local minimum. Sign intelligibility was measured for 32 experienced ASL signers who viewed individual signs. For full gray-scale dynamic signs activity-index subsampling yielded sequences that were significantly more intelligible than when every mth frame was chosen. This result was even more pronounced for static images. For binary images, the relative advantage of activity subsampling was smaller. We conclude that event boundaries can be defined computationally and that subsampling from event boundaries is better than choosing at regular intervals.

UR - http://www.scopus.com/inward/record.url?scp=0025425369&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025425369&partnerID=8YFLogxK

M3 - Article

VL - 16

SP - 282

EP - 294

JO - Journal of Experimental Psychology: Human Perception and Performance

JF - Journal of Experimental Psychology: Human Perception and Performance

SN - 0096-1523

IS - 2

ER -