Comparative Analysis: Violence Recognition from Videos using Transfer Learning

TL;DR

This paper benchmarks various deep learning models for violence recognition in videos, demonstrating that increasing dataset size from 500 to 1,600 videos improves accuracy by 6%, highlighting the importance of data volume.

Contribution

It provides a comparative analysis of deep learning techniques for violence detection and evaluates the impact of larger datasets on model performance.

Findings

Accuracy improved by 6% with more data

Deep learning models vary in effectiveness

Larger datasets enhance violence recognition

Abstract

Action recognition has become a hot topic in computer vision. However, the main applications of computer vision in video processing have focused on detection of relatively simple actions while complex events such as violence detection have been comparatively less investigated. This study focuses on the benchmarking of various deep learning techniques on a complex dataset. Next, a larger dataset is utilized to test the uplift from increasing volume of data. The dataset size increase from 500 to 1,600 videos resulted in a notable average accuracy improvement of 6% across four models.