Salvato in:
Dettagli Bibliografici
Autore:Martinel, Niki
Titolo:A Distributed Video Surveillance System to Track Persons in Camera Networks
Pubblicazione:: Università degli Studi di Udine, 2014-04-09
Abstract: This thesis focuses on the topics of information visualization in a video surveillance system and on distributed person re-identification. Visualizing the proper information in a concise and informative fashion is a very challenging tasks that is ma...distributed person re-identification. Visualizing the proper information in a concise and informative fashion is a very challenging tasks that is mainly driven by the situation, the task that has to be performed and last but not least, by the operator that is using the system. Designing a successful system that is able to support all of these requirements and constraints is the first goal of this thesis. Towards this end we design and develop four system prototypes and evaluate each of them by means of standard Human-Computer Interaction principles. We show that this approach leads to an advanced system that is capable to support the task of tracking a person moving through multiple cameras field-of-views using only a single display. The advanced visualization system was built to support the task of tracking a person through multiple overlapping cameras. However, in a real scenario this is not always feasible and we have to deal with disjoint cameras, hence, the system may fail to track the same person moving across them. In light of this, we propose three different methods to tackle the person re-identification problem so as we can re-associate a person that moves out from one camera and then reappears in another one at a different time instant. The first method builds a discriminative signature for each person that is matched by using a robust distance measure. The second method studies the transformation of features across cameras, while the last one builds upon the idea that as features get transformed so is the distance between them. Finally, we consider the issues of a fully centralized camera-camera re-identification system and introduce a distributed re-identification framework. For each approach, experimental results on public benchmark datasets are given.
Read more
Note:diritti: info:eu-repo/semantics/openAccess
In relazione con info:eu-repo/semantics/altIdentifier/hdl/11390/1132649
Autori secondari:MICHELONI, CHRISTIAN
Classe MIUR:Settore INF/01 - - Informatica
Risorsa digitale:Copia depositata in BNCF Repository di Ateneo
Copia depositata in BNCF Repository di Ateneo
LEADER 03087cam a2200277 n 4500
001 TD18003763
005 20201209081901.0
049 |a TDMAGDIG 
100 |a 20190501d2014 --k--ita-50----ba  
101 1 |a eng 
200 1 |a A Distributed Video Surveillance System to Track Persons in Camera Networks  |b Tesi di dottorato 
210 1 |c Università degli Studi di Udine  |d 2014-04-09 
300 |a diritti: info:eu-repo/semantics/openAccess 
300 |a In relazione con info:eu-repo/semantics/altIdentifier/hdl/11390/1132649 
328 0 |b tesi di dottorato  |c Settore INF/01 - Informatica  |e Università degli Studi di Udine 
330 |a This thesis focuses on the topics of information visualization in a video surveillance system and on distributed person re-identification. Visualizing the proper information in a concise and informative fashion is a very challenging tasks that is mainly driven by the situation, the task that has to be performed and last but not least, by the operator that is using the system. Designing a successful system that is able to support all of these requirements and constraints is the first goal of this thesis. Towards this end we design and develop four system prototypes and evaluate each of them by means of standard Human-Computer Interaction principles. We show that this approach leads to an advanced system that is capable to support the task of tracking a person moving through multiple cameras field-of-views using only a single display. The advanced visualization system was built to support the task of tracking a person through multiple overlapping cameras. However, in a real scenario this is not always feasible and we have to deal with disjoint cameras, hence, the system may fail to track the same person moving across them. In light of this, we propose three different methods to tackle the person re-identification problem so as we can re-associate a person that moves out from one camera and then reappears in another one at a different time instant. The first method builds a discriminative signature for each person that is matched by using a robust distance measure. The second method studies the transformation of features across cameras, while the last one builds upon the idea that as features get transformed so is the distance between them. Finally, we consider the issues of a fully centralized camera-camera re-identification system and introduce a distributed re-identification framework. For each approach, experimental results on public benchmark datasets are given. 
689 0 |a Settore INF/01  |b - Informatica  |c TDR 
700 0 |a Martinel, Niki 
702 0 |a MICHELONI, CHRISTIAN 
801 3 |a IT  |b IT-FI0098 
856 4 |u http://memoria.depositolegale.it/*/http://hdl.handle.net/11390/1132649  |2 http://hdl.handle.net/11390/1132649 
856 4 |u http://memoria.depositolegale.it/*/http://air.uniud.it/bitstream/11390/1132649/1/10990_447_PhD_Thesis_Niki_Martinel.pdf  |2 http://air.uniud.it/bitstream/11390/1132649/1/10990_447_PhD_Thesis_Niki_Martinel.pdf 
977 |a  CR 
997 |a CF 
FMT |a TD 
FOR |a TD 
Tesi di dottorato | Lingua: Inglese | Paese: | BID: TD18003763
Visualizza in formato Unimarc
Biblioteca Inventario Volume Collocazione Fruizione status