STAC: Leveraging Spatio-Temporal Data Associations For Efficient Cross-Camera Streaming and Analytics

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Gupta, Ragini, Zhao, Lingzhi, Li, Jiaxi, Vakhniuk, Volodymyr, Danilov, Claudiu, Eckhardt, Josh, Bernard, Keyshla, Nahrstedt, Klara
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912536091164672
author Gupta, Ragini
Zhao, Lingzhi
Li, Jiaxi
Vakhniuk, Volodymyr
Danilov, Claudiu
Eckhardt, Josh
Bernard, Keyshla
Nahrstedt, Klara
author_facet Gupta, Ragini
Zhao, Lingzhi
Li, Jiaxi
Vakhniuk, Volodymyr
Danilov, Claudiu
Eckhardt, Josh
Bernard, Keyshla
Nahrstedt, Klara
contents In IoT based distributed network of cameras, real-time multi-camera video analytics is challenged by high bandwidth demands and redundant visual data, creating a fundamental tension where reducing data saves network overhead but can degrade model performance, and vice versa. We present STAC, a cross-cameras surveillance system that leverages spatio-temporal associations for efficient object tracking under constrained network conditions. STAC integrates multi-resolution feature learning, ensuring robustness under variable networked system level optimizations such as frame filtering, FFmpeg-based compression, and Region-of-Interest (RoI) masking, to eliminate redundant content across distributed video streams while preserving downstream model accuracy for object identification and tracking. Evaluated on NVIDIA's AICity Challenge dataset, STAC achieves a 76\% improvement in tracking accuracy and an 8.6x reduction in inference latency over a standard multi-object multi-camera tracking baseline (using YOLOv4 and DeepSORT). Furthermore, 29\% of redundant frames are filtered, significantly reducing data volume without compromising inference quality.
format Preprint
id arxiv_https___arxiv_org_abs_2401_15288
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle STAC: Leveraging Spatio-Temporal Data Associations For Efficient Cross-Camera Streaming and Analytics
Gupta, Ragini
Zhao, Lingzhi
Li, Jiaxi
Vakhniuk, Volodymyr
Danilov, Claudiu
Eckhardt, Josh
Bernard, Keyshla
Nahrstedt, Klara
Computer Vision and Pattern Recognition
Multimedia
Networking and Internet Architecture
I.4.2; I.4.0; C.2.2; C.2.0
In IoT based distributed network of cameras, real-time multi-camera video analytics is challenged by high bandwidth demands and redundant visual data, creating a fundamental tension where reducing data saves network overhead but can degrade model performance, and vice versa. We present STAC, a cross-cameras surveillance system that leverages spatio-temporal associations for efficient object tracking under constrained network conditions. STAC integrates multi-resolution feature learning, ensuring robustness under variable networked system level optimizations such as frame filtering, FFmpeg-based compression, and Region-of-Interest (RoI) masking, to eliminate redundant content across distributed video streams while preserving downstream model accuracy for object identification and tracking. Evaluated on NVIDIA's AICity Challenge dataset, STAC achieves a 76\% improvement in tracking accuracy and an 8.6x reduction in inference latency over a standard multi-object multi-camera tracking baseline (using YOLOv4 and DeepSORT). Furthermore, 29\% of redundant frames are filtered, significantly reducing data volume without compromising inference quality.
title STAC: Leveraging Spatio-Temporal Data Associations For Efficient Cross-Camera Streaming and Analytics
topic Computer Vision and Pattern Recognition
Multimedia
Networking and Internet Architecture
I.4.2; I.4.0; C.2.2; C.2.0
url https://arxiv.org/abs/2401.15288