Reducto: On-Camera Filtering for Resource-Efficient Real-Time Video Analytics

Yuanqi Li, Arthi Padmanabhan, Pengzhan Zhao, Yufei Wang, Guoqing Harry Xu, Ravi Netravali

Research output: Chapter in Book/Report/Conference proceedingConference contribution

85 Scopus citations

Abstract

To cope with the high resource (network and compute) demands of real-time video analytics pipelines, recent systems have relied on frame filtering. However, filtering has typically been done with neural networks running on edge/backend servers that are expensive to operate. This paper investigates on-camera filtering, which moves filtering to the beginning of the pipeline. Unfortunately, we find that commodity cameras have limited compute resources that only permit filtering via frame differencing based on low-level video features. Used incorrectly, such techniques can lead to unacceptable drops in query accuracy. To overcome this, we built Reducto, a system that dynamically adapts filtering decisions according to the time-varying correlation between feature type, filtering threshold, query accuracy, and video content. Experiments with a variety of videos and queries show that Reducto achieves significant (51-97% of frames) filtering benefits, while consistently meeting the desired accuracy.

Original languageEnglish (US)
Title of host publicationSIGCOMM 2020 - Proceedings of the 2020 Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, Technologies, Architectures, and Protocols for Computer Communication
PublisherAssociation for Computing Machinery
Pages359-376
Number of pages18
ISBN (Electronic)9781450379557
DOIs
StatePublished - Jul 30 2020
Externally publishedYes
Event2020 Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, Technologies, Architectures, and Protocols for Computer Communication, SIGCOMM 2020 - Virtual, Online, United States
Duration: Aug 10 2020Aug 14 2020

Publication series

NameSIGCOMM 2020 - Proceedings of the 2020 Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, Technologies, Architectures, and Protocols for Computer Communication

Conference

Conference2020 Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, Technologies, Architectures, and Protocols for Computer Communication, SIGCOMM 2020
Country/TerritoryUnited States
CityVirtual, Online
Period8/10/208/14/20

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Keywords

  • deep neural networks
  • object detection
  • video analytics

Fingerprint

Dive into the research topics of 'Reducto: On-Camera Filtering for Resource-Efficient Real-Time Video Analytics'. Together they form a unique fingerprint.

Cite this