Omni-directional motion: pedestrian shape classification using neural networks and active contour models

Ken Tabb; S. George; N. Davey; R.G. Adams

Omni-directional motion: pedestrian shape classification using neural networks and active contour models

Ken Tabb, S. George, N. Davey, R.G. Adams

Centre for Computer Science and Informatics Research

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

35 Downloads (Pure)

Abstract

This paper describes a hybrid vision system which, following initial user
interaction, can detect and track objects in the visual field, and classify them as human and non-human. The system incorporates an active contour model for detecting and tracking objects, a method of translating the contours into scale-, location- and resolution-independent vectors, and an error-backpropagation feedforward neural network for shape classification of these vectors. The network is able to generate a confidence value for a given shape, determining how ‘human’ and how ‘non-human’ it considers the shape to be. This confidence value changes as the object moves around, providing a motion signature for an object. Previous work has accommodated lateral pedestrian movement across the visual field; this paper describes a system which accommodates all angles of pedestrian movement on the ground plane.

Original language	English
Title of host publication	In: Proc. of Image and Vision Computing New Zealand (IVCNZ)
Pages	387-392
Publication status	Published - 2001
Event	Image and Vision Computing New Zealand (IVCNZ) - Dunedin, New Zealand Duration: 26 Nov 2001 → 28 Nov 2001

Conference

Conference	Image and Vision Computing New Zealand (IVCNZ)
Country/Territory	New Zealand
City	Dunedin
Period	26/11/01 → 28/11/01

Keywords

snake
active contour model
shape classification
neural network
Omni-directional
Axis crossover vector
ground plane

Access to Document

900897.pdf

Cite this

@inproceedings{9e628d8cc75b4435a7f4a30753ae0e28,

title = "Omni-directional motion: pedestrian shape classification using neural networks and active contour models",

abstract = "This paper describes a hybrid vision system which, following initial user interaction, can detect and track objects in the visual field, and classify them as human and non-human. The system incorporates an active contour model for detecting and tracking objects, a method of translating the contours into scale-, location- and resolution-independent vectors, and an error-backpropagation feedforward neural network for shape classification of these vectors. The network is able to generate a confidence value for a given shape, determining how {\textquoteleft}human{\textquoteright} and how {\textquoteleft}non-human{\textquoteright} it considers the shape to be. This confidence value changes as the object moves around, providing a motion signature for an object. Previous work has accommodated lateral pedestrian movement across the visual field; this paper describes a system which accommodates all angles of pedestrian movement on the ground plane.",

keywords = "snake, active contour model, shape classification, neural network, Omni-directional, Axis crossover vector, ground plane",

author = "Ken Tabb and S. George and N. Davey and R.G. Adams",

year = "2001",

language = "English",

pages = "387--392",

booktitle = "In: Proc. of Image and Vision Computing New Zealand (IVCNZ)",

note = "Image and Vision Computing New Zealand (IVCNZ) ; Conference date: 26-11-2001 Through 28-11-2001",

}

TY - GEN

T1 - Omni-directional motion: pedestrian shape classification using neural networks and active contour models

AU - Tabb, Ken

AU - George, S.

AU - Davey, N.

AU - Adams, R.G.

PY - 2001

Y1 - 2001

N2 - This paper describes a hybrid vision system which, following initial user interaction, can detect and track objects in the visual field, and classify them as human and non-human. The system incorporates an active contour model for detecting and tracking objects, a method of translating the contours into scale-, location- and resolution-independent vectors, and an error-backpropagation feedforward neural network for shape classification of these vectors. The network is able to generate a confidence value for a given shape, determining how ‘human’ and how ‘non-human’ it considers the shape to be. This confidence value changes as the object moves around, providing a motion signature for an object. Previous work has accommodated lateral pedestrian movement across the visual field; this paper describes a system which accommodates all angles of pedestrian movement on the ground plane.

AB - This paper describes a hybrid vision system which, following initial user interaction, can detect and track objects in the visual field, and classify them as human and non-human. The system incorporates an active contour model for detecting and tracking objects, a method of translating the contours into scale-, location- and resolution-independent vectors, and an error-backpropagation feedforward neural network for shape classification of these vectors. The network is able to generate a confidence value for a given shape, determining how ‘human’ and how ‘non-human’ it considers the shape to be. This confidence value changes as the object moves around, providing a motion signature for an object. Previous work has accommodated lateral pedestrian movement across the visual field; this paper describes a system which accommodates all angles of pedestrian movement on the ground plane.

KW - snake

KW - active contour model

KW - shape classification

KW - neural network

KW - Omni-directional

KW - Axis crossover vector

KW - ground plane

M3 - Conference contribution

SP - 387

EP - 392

BT - In: Proc. of Image and Vision Computing New Zealand (IVCNZ)

T2 - Image and Vision Computing New Zealand (IVCNZ)

Y2 - 26 November 2001 through 28 November 2001

ER -

Omni-directional motion: pedestrian shape classification using neural networks and active contour models

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this