2021
Durães, D.; Santos, F.; Marcondes, F. S.; Lange, S.; Machado, J.
Comparison of Transfer Learning Behaviour in Violence Detection with Different Public Datasets Proceedings Article
Em: G., Lau N. Melo F. S. Marreiros (Ed.): pp. 290-298, Springer Science and Business Media Deutschland GmbH, 2021, ISSN: 03029743, (cited By 3; Conference of 20th EPIA Conference on Artificial Intelligence, EPIA 2021 ; Conference Date: 7 September 2021 Through 9 September 2021; Conference Code:265179).
Resumo | Links | BibTeX | Etiquetas: Area of interest; Deep learning; Inside car; Learning behavior; Pre-training; Public dataset; Real situation; Real- time; Video recognition; Violence detections, Deep learning; Human computer interaction, Security systems
@inproceedings{Durães2021290,
title = {Comparison of Transfer Learning Behaviour in Violence Detection with Different Public Datasets},
author = {D. Durães and F. Santos and F. S. Marcondes and S. Lange and J. Machado},
editor = {Lau N. Melo F.S. Marreiros G.},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85115443718&doi=10.1007%2f978-3-030-86230-5_23&partnerID=40&md5=258a2451e74014f8e5df8c461bf232a0},
doi = {10.1007/978-3-030-86230-5_23},
issn = {03029743},
year = {2021},
date = {2021-01-01},
journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
volume = {12981 LNAI},
pages = {290-298},
publisher = {Springer Science and Business Media Deutschland GmbH},
abstract = {The detection and recognition of violence have been area of interest to research, mainly in surveillance, Human-Computer Interaction and information retrieval for video based on content. The primary purpose of detecting and recognizing violence is to automatically and in real-time recognize violence. Hence, it is a crucial area and object of several studies, as it will enable systems to have the necessary means to contain violence automatically. In this sense, pre-trained models are used to solve general problems of recognition of violent activity. These models were pre-trained with datasets from: hockey fight; movies; violence in real surveillance; and fighting in real situations. From this pre-training models, general patterns are extracted that are very important to detect violent behaviour in videos. Our approach uses a state-of-the-art pre-trained violence detection model in general activity recognition tasks and then tweaks it for violence detection inside a car. For this, we created our dataset with videos inside the car to apply in this study. © 2021, Springer Nature Switzerland AG.},
note = {cited By 3; Conference of 20th EPIA Conference on Artificial Intelligence, EPIA 2021 ; Conference Date: 7 September 2021 Through 9 September 2021; Conference Code:265179},
keywords = {Area of interest; Deep learning; Inside car; Learning behavior; Pre-training; Public dataset; Real situation; Real- time; Video recognition; Violence detections, Deep learning; Human computer interaction, Security systems},
pubstate = {published},
tppubtype = {inproceedings}
}
The detection and recognition of violence have been area of interest to research, mainly in surveillance, Human-Computer Interaction and information retrieval for video based on content. The primary purpose of detecting and recognizing violence is to automatically and in real-time recognize violence. Hence, it is a crucial area and object of several studies, as it will enable systems to have the necessary means to contain violence automatically. In this sense, pre-trained models are used to solve general problems of recognition of violent activity. These models were pre-trained with datasets from: hockey fight; movies; violence in real surveillance; and fighting in real situations. From this pre-training models, general patterns are extracted that are very important to detect violent behaviour in videos. Our approach uses a state-of-the-art pre-trained violence detection model in general activity recognition tasks and then tweaks it for violence detection inside a car. For this, we created our dataset with videos inside the car to apply in this study. © 2021, Springer Nature Switzerland AG.