Identification of Social-Media Platform of Videos through the Use of Shared Features

TL;DR

This paper introduces transfer learning and multitask learning methods to identify the social media platform of videos using shared features with images, addressing privacy constraints and demonstrating the effectiveness of the multitask approach.

Contribution

It presents the first approach to identify social media platforms of videos using shared features and compares transfer learning with multitask learning techniques.

Findings

Multitask learning outperforms transfer learning in platform identification.

Shared features between images and videos are effective for this classification.

The proposed methods show promising experimental results.

Abstract

Videos have become a powerful tool for spreading illegal content such as military propaganda, revenge porn, or bullying through social networks. To counter these illegal activities, it has become essential to try new methods to verify the origin of videos from these platforms. However, collecting datasets large enough to train neural networks for this task has become difficult because of the privacy regulations that have been enacted in recent years. To mitigate this limitation, in this work we propose two different solutions based on transfer learning and multitask learning to determine whether a video has been uploaded from or downloaded to a specific social platform through the use of shared features with images trained on the same task. By transferring features from the shallowest to the deepest levels of the network from the image task to videos, we measure the amount of information shared between these two tasks. Then, we introduce a model based on multitask learning, which learns from both tasks simultaneously. The promising experimental results show, in particular, the effectiveness of the multitask approach. According to our knowledge, this is the first work that addresses the problem of social media platform identification of videos through the use of shared features.