Guardado en:
Detalles Bibliográficos
Autores principales: Li, Xiwen, Tang, Xiaoya, Tasdizen, Tolga
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2504.16102
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908608099254272
author Li, Xiwen
Tang, Xiaoya
Tasdizen, Tolga
author_facet Li, Xiwen
Tang, Xiaoya
Tasdizen, Tolga
contents Idling vehicle detection (IVD) uses surveillance video and multichannel audio to localize and classify vehicles in the last frame as moving, idling, or engine-off in pick-up zones. IVD faces three challenges: (i) modality heterogeneity between visual cues and audio patterns; (ii) large box scale variation requiring multi-resolution detection; and (iii) training instability due to coupled detection heads. The previous end-to-end (E2E) model with simple CBAM-based bi-modal attention fails to handle these issues and often misses vehicles. We propose HAVT-IVD, a heterogeneity-aware network with a visual feature pyramid and decoupled heads. Experiments show HAVT-IVD improves mAP by 7.66 over the disjoint baseline and 9.42 over the E2E baseline.
format Preprint
id arxiv_https___arxiv_org_abs_2504_16102
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HAVT-IVD: Heterogeneity-Aware Cross-Modal Network for Audio-Visual Surveillance: Idling Vehicles Detection With Multichannel Audio and Multiscale Visual Cues
Li, Xiwen
Tang, Xiaoya
Tasdizen, Tolga
Computer Vision and Pattern Recognition
Robotics
Idling vehicle detection (IVD) uses surveillance video and multichannel audio to localize and classify vehicles in the last frame as moving, idling, or engine-off in pick-up zones. IVD faces three challenges: (i) modality heterogeneity between visual cues and audio patterns; (ii) large box scale variation requiring multi-resolution detection; and (iii) training instability due to coupled detection heads. The previous end-to-end (E2E) model with simple CBAM-based bi-modal attention fails to handle these issues and often misses vehicles. We propose HAVT-IVD, a heterogeneity-aware network with a visual feature pyramid and decoupled heads. Experiments show HAVT-IVD improves mAP by 7.66 over the disjoint baseline and 9.42 over the E2E baseline.
title HAVT-IVD: Heterogeneity-Aware Cross-Modal Network for Audio-Visual Surveillance: Idling Vehicles Detection With Multichannel Audio and Multiscale Visual Cues
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2504.16102