Fight Detection from Still Images in the Wild

TL;DR

This paper investigates the feasibility of detecting fights from single still images shared on social media, introduces a new dataset, and demonstrates high accuracy in fight recognition without temporal data, highlighting dataset biases.

Contribution

The study presents a new real-world fight image dataset (SMFI) and analyzes fight detection accuracy from still images, revealing dataset bias issues in fight recognition.

Findings

Fight detection from still images is feasible with high accuracy.

The proposed SMFI dataset is highly representative among existing datasets.

Cross-dataset experiments show significant accuracy drop, indicating dataset bias.

Abstract

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.