A deep-learning approach to early identification of suggested sexual harassment from videos

TL;DR

This paper introduces a deep learning-based method for early detection of sexual harassment in videos, utilizing a new annotated dataset from Indian movies to improve classification accuracy.

Contribution

It presents a novel dataset of movie scenes labeled for harassment, abuse, and violence, and evaluates the limitations of existing explicit content detectors.

Findings

State-of-the-art detectors fail to identify harassment scenes effectively.

The new dataset enables training of specialized deep learning models.

Facial expressions and unwanted touching are key indicators.

Abstract

Sexual harassment, sexual abuse, and sexual violence are prevalent problems in this day and age. Women's safety is an important issue that needs to be highlighted and addressed. Given this issue, we have studied each of these concerns and the factors that affect it based on images generated from movies. We have classified the three terms (harassment, abuse, and violence) based on the visual attributes present in images depicting these situations. We identified that factors such as facial expression of the victim and perpetrator and unwanted touching had a direct link to identifying the scenes containing sexual harassment, abuse and violence. We also studied and outlined how state-of-the-art explicit content detectors such as Google Cloud Vision API and Clarifai API fail to identify and categorise these images. Based on these definitions and characteristics, we have developed a first-of-its-kind dataset from various Indian movie scenes. These scenes are classified as sexual harassment, sexual abuse, or sexual violence and exported in the PASCAL VOC 1.1 format. Our dataset is annotated on the identified relevant features and can be used to develop and train a deep-learning computer vision model to identify these issues. The dataset is publicly available for research and development.