Motion analysis by feature tracking

Michela Del Viva, M. Concetta Morrone

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

We have developed a two-stage model of motion perception that identifies moving spatial features and computes their velocity, achieving both high spatial localisation and reliable estimates of velocity. Features are detected in each frame by locating the peaks of the spatial local energy functions, as for stationary images. The energy functions are calculated for different scales and orientations, and integrated within a temporal Gaussian window. The velocity of features is determined by the direction of maximal elongation of the energy in space-time, evaluated by calculating the three characteristic curvatures of the energy at each feature point. To circumvent the aperture problem, the energy maps are blurred in space by various amounts, and velocity is computed separately for each spatial blur. The Weber fraction of the local curvatures (curvature contrast) describes the spatio-temporal energy elongation at each feature point, giving a reliability index for each velocity estimate. For each point, the velocity of the spatial blur that yielded the highest curvature contrast was selected, with no further constraints, such as rigidity of motion. Dynamic recruitment of operators of different size allows maximum flexibility of the analysis, allowing it to simulate human visual performance in the detection of noise images, transparent motion, some motion illusions, and second-order motion.

Original languageEnglish
Pages (from-to)3633-3653
Number of pages21
JournalVision Research
Volume38
Issue number22
DOIs
Publication statusPublished - Nov 1998

Fingerprint

Motion Perception
Noise
Direction compound

Keywords

  • Feature tracking
  • Motion modelling
  • Motion transparency
  • Second-order motion
  • Spatio-temporal filters

ASJC Scopus subject areas

  • Ophthalmology
  • Sensory Systems

Cite this

Motion analysis by feature tracking. / Del Viva, Michela; Morrone, M. Concetta.

In: Vision Research, Vol. 38, No. 22, 11.1998, p. 3633-3653.

Research output: Contribution to journalArticle

Del Viva, Michela ; Morrone, M. Concetta. / Motion analysis by feature tracking. In: Vision Research. 1998 ; Vol. 38, No. 22. pp. 3633-3653.
@article{fc1fd2b799fd4609986a04dc3d3cbf0e,
title = "Motion analysis by feature tracking",
abstract = "We have developed a two-stage model of motion perception that identifies moving spatial features and computes their velocity, achieving both high spatial localisation and reliable estimates of velocity. Features are detected in each frame by locating the peaks of the spatial local energy functions, as for stationary images. The energy functions are calculated for different scales and orientations, and integrated within a temporal Gaussian window. The velocity of features is determined by the direction of maximal elongation of the energy in space-time, evaluated by calculating the three characteristic curvatures of the energy at each feature point. To circumvent the aperture problem, the energy maps are blurred in space by various amounts, and velocity is computed separately for each spatial blur. The Weber fraction of the local curvatures (curvature contrast) describes the spatio-temporal energy elongation at each feature point, giving a reliability index for each velocity estimate. For each point, the velocity of the spatial blur that yielded the highest curvature contrast was selected, with no further constraints, such as rigidity of motion. Dynamic recruitment of operators of different size allows maximum flexibility of the analysis, allowing it to simulate human visual performance in the detection of noise images, transparent motion, some motion illusions, and second-order motion.",
keywords = "Feature tracking, Motion modelling, Motion transparency, Second-order motion, Spatio-temporal filters",
author = "{Del Viva}, Michela and Morrone, {M. Concetta}",
year = "1998",
month = "11",
doi = "10.1016/S0042-6989(98)00022-4",
language = "English",
volume = "38",
pages = "3633--3653",
journal = "Vision Research",
issn = "0042-6989",
publisher = "Elsevier Limited",
number = "22",

}

TY - JOUR

T1 - Motion analysis by feature tracking

AU - Del Viva, Michela

AU - Morrone, M. Concetta

PY - 1998/11

Y1 - 1998/11

N2 - We have developed a two-stage model of motion perception that identifies moving spatial features and computes their velocity, achieving both high spatial localisation and reliable estimates of velocity. Features are detected in each frame by locating the peaks of the spatial local energy functions, as for stationary images. The energy functions are calculated for different scales and orientations, and integrated within a temporal Gaussian window. The velocity of features is determined by the direction of maximal elongation of the energy in space-time, evaluated by calculating the three characteristic curvatures of the energy at each feature point. To circumvent the aperture problem, the energy maps are blurred in space by various amounts, and velocity is computed separately for each spatial blur. The Weber fraction of the local curvatures (curvature contrast) describes the spatio-temporal energy elongation at each feature point, giving a reliability index for each velocity estimate. For each point, the velocity of the spatial blur that yielded the highest curvature contrast was selected, with no further constraints, such as rigidity of motion. Dynamic recruitment of operators of different size allows maximum flexibility of the analysis, allowing it to simulate human visual performance in the detection of noise images, transparent motion, some motion illusions, and second-order motion.

AB - We have developed a two-stage model of motion perception that identifies moving spatial features and computes their velocity, achieving both high spatial localisation and reliable estimates of velocity. Features are detected in each frame by locating the peaks of the spatial local energy functions, as for stationary images. The energy functions are calculated for different scales and orientations, and integrated within a temporal Gaussian window. The velocity of features is determined by the direction of maximal elongation of the energy in space-time, evaluated by calculating the three characteristic curvatures of the energy at each feature point. To circumvent the aperture problem, the energy maps are blurred in space by various amounts, and velocity is computed separately for each spatial blur. The Weber fraction of the local curvatures (curvature contrast) describes the spatio-temporal energy elongation at each feature point, giving a reliability index for each velocity estimate. For each point, the velocity of the spatial blur that yielded the highest curvature contrast was selected, with no further constraints, such as rigidity of motion. Dynamic recruitment of operators of different size allows maximum flexibility of the analysis, allowing it to simulate human visual performance in the detection of noise images, transparent motion, some motion illusions, and second-order motion.

KW - Feature tracking

KW - Motion modelling

KW - Motion transparency

KW - Second-order motion

KW - Spatio-temporal filters

UR - http://www.scopus.com/inward/record.url?scp=0031719611&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031719611&partnerID=8YFLogxK

U2 - 10.1016/S0042-6989(98)00022-4

DO - 10.1016/S0042-6989(98)00022-4

M3 - Article

C2 - 9893795

AN - SCOPUS:0031719611

VL - 38

SP - 3633

EP - 3653

JO - Vision Research

JF - Vision Research

SN - 0042-6989

IS - 22

ER -