Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data

Csaba Beleznai, Michael Rauter, Martin Hirzer, Peter M. Roth

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Visual surveillance data might encompass vast data amounts. Given the amount of data the need for search and data exploration arises naturally. Various authorities such as infrastructure operators and law enforcement agencies are confronted with search needs based on a visual description and/or behavioral patterns (motion path, activity) in order to find a ”needle in a haystack of digital data”. In this paper we present a framework which allows for an efficient search in visual surveillance archives. The paper describes following core algorithmic components of the search framework: Human detection employing pedestrian-specific shape and motion cues along with occlusion modelling; Tracking of multiple interacting pedestrians using a hierarchical spatio-temporal association scheme. Finally, pedestrian re-identification is demonstrated based on appearance matching in order to recognize a given person across a network of spatially disjoint cameras. We present results for the detection, tracking and re-identification subtasks on various challenging datasets and describe the overall framework in detail.
LanguageEnglish
Title of host publicationProceedings of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF)
Publisher.
StatusPublished - 2013

Fingerprint

Law enforcement
Cameras

Fields of Expertise

  • Information, Communication & Computing

Cite this

Beleznai, C., Rauter, M., Hirzer, M., & Roth, P. M. (2013). Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data. In Proceedings of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF) ..

Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data. / Beleznai, Csaba; Rauter, Michael; Hirzer, Martin; Roth, Peter M.

Proceedings of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF). ., 2013.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Beleznai, C, Rauter, M, Hirzer, M & Roth, PM 2013, Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data. in Proceedings of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF). .
Beleznai C, Rauter M, Hirzer M, Roth PM. Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data. In Proceedings of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF). . 2013
Beleznai, Csaba ; Rauter, Michael ; Hirzer, Martin ; Roth, Peter M. / Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data. Proceedings of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF). ., 2013.
@inproceedings{b27c666db9a242ffb93c8d84351565da,
title = "Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data",
abstract = "Visual surveillance data might encompass vast data amounts. Given the amount of data the need for search and data exploration arises naturally. Various authorities such as infrastructure operators and law enforcement agencies are confronted with search needs based on a visual description and/or behavioral patterns (motion path, activity) in order to find a ”needle in a haystack of digital data”. In this paper we present a framework which allows for an efficient search in visual surveillance archives. The paper describes following core algorithmic components of the search framework: Human detection employing pedestrian-specific shape and motion cues along with occlusion modelling; Tracking of multiple interacting pedestrians using a hierarchical spatio-temporal association scheme. Finally, pedestrian re-identification is demonstrated based on appearance matching in order to recognize a given person across a network of spatially disjoint cameras. We present results for the detection, tracking and re-identification subtasks on various challenging datasets and describe the overall framework in detail.",
author = "Csaba Beleznai and Michael Rauter and Martin Hirzer and Roth, {Peter M.}",
year = "2013",
language = "English",
booktitle = "Proceedings of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF)",
publisher = ".",

}

TY - GEN

T1 - Pedestrian Detection, Tracking and Re-Identification for Search in Visual Surveillance Data

AU - Beleznai, Csaba

AU - Rauter, Michael

AU - Hirzer, Martin

AU - Roth, Peter M.

PY - 2013

Y1 - 2013

N2 - Visual surveillance data might encompass vast data amounts. Given the amount of data the need for search and data exploration arises naturally. Various authorities such as infrastructure operators and law enforcement agencies are confronted with search needs based on a visual description and/or behavioral patterns (motion path, activity) in order to find a ”needle in a haystack of digital data”. In this paper we present a framework which allows for an efficient search in visual surveillance archives. The paper describes following core algorithmic components of the search framework: Human detection employing pedestrian-specific shape and motion cues along with occlusion modelling; Tracking of multiple interacting pedestrians using a hierarchical spatio-temporal association scheme. Finally, pedestrian re-identification is demonstrated based on appearance matching in order to recognize a given person across a network of spatially disjoint cameras. We present results for the detection, tracking and re-identification subtasks on various challenging datasets and describe the overall framework in detail.

AB - Visual surveillance data might encompass vast data amounts. Given the amount of data the need for search and data exploration arises naturally. Various authorities such as infrastructure operators and law enforcement agencies are confronted with search needs based on a visual description and/or behavioral patterns (motion path, activity) in order to find a ”needle in a haystack of digital data”. In this paper we present a framework which allows for an efficient search in visual surveillance archives. The paper describes following core algorithmic components of the search framework: Human detection employing pedestrian-specific shape and motion cues along with occlusion modelling; Tracking of multiple interacting pedestrians using a hierarchical spatio-temporal association scheme. Finally, pedestrian re-identification is demonstrated based on appearance matching in order to recognize a given person across a network of spatially disjoint cameras. We present results for the detection, tracking and re-identification subtasks on various challenging datasets and describe the overall framework in detail.

M3 - Conference contribution

BT - Proceedings of the Hungarian Association for Image Processing and Pattern Recognition (KEPAF)

PB - .

ER -