TY - GEN
T1 - Fight Detection from Still Images in the Wild
AU - Akti, Seymanur
AU - Ofli, Ferda
AU - Imran, Muhammad
AU - Ekenel, Hazim Kemal
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Detecting fights from still images shared on social media is an important task required to limit the distribution of violent scenes in order to prevent their negative effects. For this reason, in this study, we address the problem of fight detection from still images collected from the web and social media. We explore how well one can detect fights from just a single still image. We also propose a new dataset, named Social Media Fight Images (SMFI), comprising real-world images of fight actions. Results of the extensive experiments on the proposed dataset show that fight actions can be recognized successfully from still images. That is, even without exploiting the temporal information, it is possible to detect fights with high accuracy by utilizing appearance only. We also perform cross-dataset experiments to evaluate the representation capacity of the collected dataset. These experiments indicate that, as in the other computer vision problems, there exists a dataset bias for the fight recognition problem. Although the methods achieve close to 100% accuracy when trained and tested on the same fight dataset, the cross-dataset accuracies are significantly lower, i.e., around 70% when more representative datasets are used for training. SMFI dataset is found to be one of the two most representative datasets among the utilized five fight datasets.
AB - Detecting fights from still images shared on social media is an important task required to limit the distribution of violent scenes in order to prevent their negative effects. For this reason, in this study, we address the problem of fight detection from still images collected from the web and social media. We explore how well one can detect fights from just a single still image. We also propose a new dataset, named Social Media Fight Images (SMFI), comprising real-world images of fight actions. Results of the extensive experiments on the proposed dataset show that fight actions can be recognized successfully from still images. That is, even without exploiting the temporal information, it is possible to detect fights with high accuracy by utilizing appearance only. We also perform cross-dataset experiments to evaluate the representation capacity of the collected dataset. These experiments indicate that, as in the other computer vision problems, there exists a dataset bias for the fight recognition problem. Although the methods achieve close to 100% accuracy when trained and tested on the same fight dataset, the cross-dataset accuracies are significantly lower, i.e., around 70% when more representative datasets are used for training. SMFI dataset is found to be one of the two most representative datasets among the utilized five fight datasets.
UR - http://www.scopus.com/inward/record.url?scp=85126799933&partnerID=8YFLogxK
U2 - 10.1109/WACVW54805.2022.00061
DO - 10.1109/WACVW54805.2022.00061
M3 - Conference contribution
AN - SCOPUS:85126799933
T3 - Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2022
SP - 550
EP - 559
BT - Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2022
Y2 - 4 January 2022 through 8 January 2022
ER -