English

Unsupervised Audio-Visual Subspace Alignment for High-Stakes Deception Detection

Computer Vision and Pattern Recognition 2021-06-22 v1 Human-Computer Interaction Machine Learning

Abstract

Automated systems that detect deception in high-stakes situations can enhance societal well-being across medical, social work, and legal domains. Existing models for detecting high-stakes deception in videos have been supervised, but labeled datasets to train models can rarely be collected for most real-world applications. To address this problem, we propose the first multimodal unsupervised transfer learning approach that detects real-world, high-stakes deception in videos without using high-stakes labels. Our subspace-alignment (SA) approach adapts audio-visual representations of deception in lab-controlled low-stakes scenarios to detect deception in real-world, high-stakes situations. Our best unsupervised SA models outperform models without SA, outperform human ability, and perform comparably to a number of existing supervised models. Our research demonstrates the potential for introducing subspace-based transfer learning to model high-stakes deception and other social behaviors in real-world contexts with a scarcity of labeled behavioral data.

Keywords

Cite

@article{arxiv.2102.03673,
  title  = {Unsupervised Audio-Visual Subspace Alignment for High-Stakes Deception Detection},
  author = {Leena Mathur and Maja J Matarić},
  journal= {arXiv preprint arXiv:2102.03673},
  year   = {2021}
}

Comments

Accepted at ICASSP 2021 \c{opyright} 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of copyrighted components of this work

R2 v1 2026-06-23T22:54:22.172Z