Cross-View Cross-Modal Unsupervised Domain Adaptation for Driver Monitoring System

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bhalla, Aditi, Hellert, Christian, Kasneci, Enkelejda
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915620401971200
author Bhalla, Aditi
Hellert, Christian
Kasneci, Enkelejda
author_facet Bhalla, Aditi
Hellert, Christian
Kasneci, Enkelejda
contents Driver distraction remains a leading cause of road traffic accidents, contributing to thousands of fatalities annually across the globe. While deep learning-based driver activity recognition methods have shown promise in detecting such distractions, their effectiveness in real-world deployments is hindered by two critical challenges: variations in camera viewpoints (cross-view) and domain shifts such as change in sensor modality or environment. Existing methods typically address either cross-view generalization or unsupervised domain adaptation in isolation, leaving a gap in the robust and scalable deployment of models across diverse vehicle configurations. In this work, we propose a novel two-phase cross-view, cross-modal unsupervised domain adaptation framework that addresses these challenges jointly on real-time driver monitoring data. In the first phase, we learn view-invariant and action-discriminative features within a single modality using contrastive learning on multi-view data. In the second phase, we perform domain adaptation to a new modality using information bottleneck loss without requiring any labeled data from the new domain. We evaluate our approach using state-of-the art video transformers (Video Swin, MViT) and multi modal driver activity dataset called Drive&Act, demonstrating that our joint framework improves top-1 accuracy on RGB video data by almost 50% compared to a supervised contrastive learning-based cross-view method, and outperforms unsupervised domain adaptation-only methods by up to 5%, using the same video transformer backbone.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12196
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Cross-View Cross-Modal Unsupervised Domain Adaptation for Driver Monitoring System
Bhalla, Aditi
Hellert, Christian
Kasneci, Enkelejda
Computer Vision and Pattern Recognition
Human-Computer Interaction
Driver distraction remains a leading cause of road traffic accidents, contributing to thousands of fatalities annually across the globe. While deep learning-based driver activity recognition methods have shown promise in detecting such distractions, their effectiveness in real-world deployments is hindered by two critical challenges: variations in camera viewpoints (cross-view) and domain shifts such as change in sensor modality or environment. Existing methods typically address either cross-view generalization or unsupervised domain adaptation in isolation, leaving a gap in the robust and scalable deployment of models across diverse vehicle configurations. In this work, we propose a novel two-phase cross-view, cross-modal unsupervised domain adaptation framework that addresses these challenges jointly on real-time driver monitoring data. In the first phase, we learn view-invariant and action-discriminative features within a single modality using contrastive learning on multi-view data. In the second phase, we perform domain adaptation to a new modality using information bottleneck loss without requiring any labeled data from the new domain. We evaluate our approach using state-of-the art video transformers (Video Swin, MViT) and multi modal driver activity dataset called Drive&Act, demonstrating that our joint framework improves top-1 accuracy on RGB video data by almost 50% compared to a supervised contrastive learning-based cross-view method, and outperforms unsupervised domain adaptation-only methods by up to 5%, using the same video transformer backbone.
title Cross-View Cross-Modal Unsupervised Domain Adaptation for Driver Monitoring System
topic Computer Vision and Pattern Recognition
Human-Computer Interaction
url https://arxiv.org/abs/2511.12196