Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis

This paper addresses the problem of markerless tracking of a human in full 3D with a high-dimensional (29D) body model. Most work in this area has been focused on achieving accurate tracking in order to replace marker-based motion capture, but do so at the cost of relying on relatively clean observi...

Full description

Bibliographic Details
Main Authors: Peursum, Patrick, Venkatesh, Svetha, West, Geoffrey
Format: Conference Paper
Published: IEEE Computer Society Conference Publishing Services 2007
Online Access:http://hdl.handle.net/20.500.11937/26678
_version_ 1848752054782132224
author Peursum, Patrick
Venkatesh, Svetha
West, Geoffrey
author_facet Peursum, Patrick
Venkatesh, Svetha
West, Geoffrey
author_sort Peursum, Patrick
building Curtin Institutional Repository
collection Online Access
description This paper addresses the problem of markerless tracking of a human in full 3D with a high-dimensional (29D) body model. Most work in this area has been focused on achieving accurate tracking in order to replace marker-based motion capture, but do so at the cost of relying on relatively clean observing conditions. This paper takes a different perspective, proposing a body-tracking model that is explicitly designed to handle real-world conditions such as occlusions by scene objects, failure recovery, long-term tracking, auto-initialisation, generalisation to different people and integration with action recognition. To achieve these goals, an action's motions are modelled with a variant of the hierarchical hidden Markov model. The model is quantitatively evaluated with several tests, including comparison to the annealed particle filter, tracking different people and tracking with a reduced resolution and frame rate.
first_indexed 2025-11-14T08:02:31Z
format Conference Paper
id curtin-20.500.11937-26678
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T08:02:31Z
publishDate 2007
publisher IEEE Computer Society Conference Publishing Services
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-266782017-01-30T12:54:43Z Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis Peursum, Patrick Venkatesh, Svetha West, Geoffrey This paper addresses the problem of markerless tracking of a human in full 3D with a high-dimensional (29D) body model. Most work in this area has been focused on achieving accurate tracking in order to replace marker-based motion capture, but do so at the cost of relying on relatively clean observing conditions. This paper takes a different perspective, proposing a body-tracking model that is explicitly designed to handle real-world conditions such as occlusions by scene objects, failure recovery, long-term tracking, auto-initialisation, generalisation to different people and integration with action recognition. To achieve these goals, an action's motions are modelled with a variant of the hierarchical hidden Markov model. The model is quantitatively evaluated with several tests, including comparison to the annealed particle filter, tracking different people and tracking with a reduced resolution and frame rate. 2007 Conference Paper http://hdl.handle.net/20.500.11937/26678 IEEE Computer Society Conference Publishing Services fulltext
spellingShingle Peursum, Patrick
Venkatesh, Svetha
West, Geoffrey
Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis
title Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis
title_full Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis
title_fullStr Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis
title_full_unstemmed Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis
title_short Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis
title_sort tracking-as-recognition for articulated full-body human motion analysis
url http://hdl.handle.net/20.500.11937/26678