Automatic Long-Term Deception Detection in Group Interaction Videos

TL;DR

This paper introduces a novel long-term deception detection framework for group interaction videos, specifically applied to the Resistance game, utilizing a new feature set called LiarRank to identify deceptive players over extended periods.

Contribution

It presents a new ensemble model and the LiarRank feature set for long-term deception detection in group videos, extending beyond single-person, short-duration analysis.

Findings

Achieved over 0.70 AUC in deception detection

Combined low-level, high-level, and LiarRank features for best performance

Demonstrated effectiveness in a complex group deception scenario

Abstract

Most work on automated deception detection (ADD) in video has two restrictions: (i) it focuses on a video of one person, and (ii) it focuses on a single act of deception in a one or two minute video. In this paper, we propose a new ADD framework which captures long term deception in a group setting. We study deception in the well-known Resistance game (like Mafia and Werewolf) which consists of 5-8 players of whom 2-3 are spies. Spies are deceptive throughout the game (typically 30-65 minutes) to keep their identity hidden. We develop an ensemble predictive model to identify spies in Resistance videos. We show that features from low-level and high-level video analysis are insufficient, but when combined with a new class of features that we call LiarRank, produce the best results. We achieve AUCs of over 0.70 in a fully automated setting. Our demo can be found at http://home.cs.dartmouth.edu/~mbolonkin/scan/demo/