Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Person tracking in 3D using Kalman filtering in single and multiple camera environments
We present a multi-camera person tracker solution that makes use of Kalman filtering principles. The tracking system could be used in conjunction with behaviour analysis systems to perform automated monitoring of human activity in a range of different environments. Targets are tracked in a 3-D world...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Department of Electrical Engineering
2024
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613244592488448 |
|---|---|
| access_status_str | Open Access |
| author | Merven, Bruno |
| author_browse | Merven, Bruno |
| author_facet | Merven, Bruno |
| author_sort | Merven, Bruno |
| collection | Thesis |
| description | We present a multi-camera person tracker solution that makes use of Kalman filtering principles. The tracking system could be used in conjunction with behaviour analysis systems to perform automated monitoring of human activity in a range of different environments. Targets are tracked in a 3-D world-view coordinate system which is common to all cameras monitoring the scene. Targets are modelled as ellipsoids and their colour information is parameterised by RGB-height histograms. Observations used to update the target models are generated by matching the targets in the different views. 3-D tracking requires that cameras are calibrated to the world coordinate system. We investigate some practical methods of obtaining this calibration information without laying out and measuring calibration markers. Both tracking and calibration methods were tested extensively using 6 different single and multiple camera test sequences. The system is able to initiate, maintain and terminate the tracks of several people in cluttered scenes. However, further optimisation of the algorithm is required to achieve tracking in real time. |
| format | Thesis |
| id | oai:open.uct.ac.za:11427/40218 |
| institution | University of Cape Town (South Africa) |
| language | eng |
| last_indexed | 2026-06-10T12:33:04.194Z |
| license_str | Not specified — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from UCTD — University of Cape Town Open Access Repository |
| publishDate | 2024 |
| publishDateRange | 2024 |
| publishDateSort | 2024 |
| publisher | Department of Electrical Engineering |
| publisherStr | Department of Electrical Engineering |
| record_format | dspace |
| source_str | UCTD — University of Cape Town Open Access Repository |
| spelling | oai:open.uct.ac.za:11427/40218 Person tracking in 3D using Kalman filtering in single and multiple camera environments Merven, Bruno Electrical Engineering We present a multi-camera person tracker solution that makes use of Kalman filtering principles. The tracking system could be used in conjunction with behaviour analysis systems to perform automated monitoring of human activity in a range of different environments. Targets are tracked in a 3-D world-view coordinate system which is common to all cameras monitoring the scene. Targets are modelled as ellipsoids and their colour information is parameterised by RGB-height histograms. Observations used to update the target models are generated by matching the targets in the different views. 3-D tracking requires that cameras are calibrated to the world coordinate system. We investigate some practical methods of obtaining this calibration information without laying out and measuring calibration markers. Both tracking and calibration methods were tested extensively using 6 different single and multiple camera test sequences. The system is able to initiate, maintain and terminate the tracks of several people in cluttered scenes. However, further optimisation of the algorithm is required to achieve tracking in real time. 2024-07-02T10:22:46Z 2024-07-02T10:22:46Z 2004 2024-06-25T13:49:56Z Thesis / Dissertation Masters MSc http://hdl.handle.net/11427/40218 eng application/pdf Department of Electrical Engineering Faculty of Engineering and the Built Environment |
| spellingShingle | Electrical Engineering Merven, Bruno Person tracking in 3D using Kalman filtering in single and multiple camera environments |
| thesis_degree_str | Master's |
| title | Person tracking in 3D using Kalman filtering in single and multiple camera environments |
| title_full | Person tracking in 3D using Kalman filtering in single and multiple camera environments |
| title_fullStr | Person tracking in 3D using Kalman filtering in single and multiple camera environments |
| title_full_unstemmed | Person tracking in 3D using Kalman filtering in single and multiple camera environments |
| title_short | Person tracking in 3D using Kalman filtering in single and multiple camera environments |
| title_sort | person tracking in 3d using kalman filtering in single and multiple camera environments |
| topic | Electrical Engineering |
| url | http://hdl.handle.net/11427/40218 |
| work_keys_str_mv | AT mervenbruno persontrackingin3dusingkalmanfilteringinsingleandmultiplecameraenvironments |