English

Scaling Video Analytics on Constrained Edge Nodes

Computer Vision and Pattern Recognition 2019-06-03 v1 Machine Learning Performance Image and Video Processing Machine Learning

Abstract

As video camera deployments continue to grow, the need to process large volumes of real-time data strains wide area network infrastructure. When per-camera bandwidth is limited, it is infeasible for applications such as traffic monitoring and pedestrian tracking to offload high-quality video streams to a datacenter. This paper presents FilterForward, a new edge-to-cloud system that enables datacenter-based applications to process content from thousands of cameras by installing lightweight edge filters that backhaul only relevant video frames. FilterForward introduces fast and expressive per-application microclassifiers that share computation to simultaneously detect dozens of events on computationally constrained edge nodes. Only matching events are transmitted to the cloud. Evaluation on two real-world camera feed datasets shows that FilterForward reduces bandwidth use by an order of magnitude while improving computational efficiency and event detection accuracy for challenging video content.

Keywords

Cite

@article{arxiv.1905.13536,
  title  = {Scaling Video Analytics on Constrained Edge Nodes},
  author = {Christopher Canel and Thomas Kim and Giulio Zhou and Conglong Li and Hyeontaek Lim and David G. Andersen and Michael Kaminsky and Subramanya R. Dulloor},
  journal= {arXiv preprint arXiv:1905.13536},
  year   = {2019}
}

Comments

This paper is an extended version of a paper with the same title published in the 2nd SysML Conference, SysML '19 (Canel et. al., 2019)

R2 v1 2026-06-23T09:34:59.836Z