direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Wissenschaftliche Veröffentlichungen

[Best Paper Award] A Local Feature based on Lagrangian Measures for Violent Video Classification
Zitatschlüssel 1479Senst2015
Autor Tobias Senst and Volker Eiselein and Thomas Sikora
Buchtitel 6th IET International Conference on Imaging for Crime Detection and Prevention
Seiten 1–6
Jahr 2015
Adresse UK, London
Monat jul
Notiz ISBN: 978-1-78561-131-5
Herausgeber Georgios Chaitas, Sergio A Velastin
Verlag IET Digital Library
Zusammenfassung Lagrangian theory provides a diverse set of tools for continuous motion analysis. Existing work shows the applicability of Lagrangian method for video analysis in several aspects. In this paper we want to utilize the concept of Lagrangian measures to detect violent scenes. Therefore we propose a local feature based on the SIFT algorithm that incooperates appearance and Lagrangian based motion models. We will show that the temporal interval of the used motion information is a crucial aspect and study its influence on the classification performance. The proposed LaSIFT feature outperforms other state-of-the-art local features, in particular in uncontrolled realistic video data. We evaluate our algorithm with a bag-of-word approach. The experimental results show a significant improvement over the state-of-the-art on current violent detection datasets, i.e. Crowd Violence, Hockey Fight.
Link zur Publikation Download Bibtex Eintrag

Zusatzinformationen / Extras


Schnellnavigation zur Seite über Nummerneingabe