Balancing Latency and Accuracy on Deep Video Analytics at the Edge

Xuebing Li, Byung Cho, Yu Xiao

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

1 Citation (Scopus)
151 Downloads (Pure)


Real-time deep video analytic at the edge is an enabling technology for emerging applications, such as vulnerable road user detection for autonomous driving, which requires highly accurate results of model inference within a low latency. In this paper, we investigate the accuracy-latency trade-off in the design and implementation of real-time deep video analytic at the edge. Without loss of generality, we select the widely used YOLO-based object detection and WebRTC-based video streaming for case study. Here, the latency consists of both networking latency caused by video streaming and the processing latency for video encoding/decoding and model inference. We conduct extensive measurements to figure out how the dynamically changing settings of video streaming affect the achieved latency, the quality of video, and further the accuracy of model inference. Based on the findings, we propose a mechanism for adapting video streaming settings (i.e. bitrate, resolution) online to optimize the accuracy of video analytic within latency constraints. The mechanism has proved, through a simulated setup, to be efficient in searching the optimal settings.
Original languageEnglish
Title of host publication2022 IEEE 19th Annual Consumer Communications & Networking Conference
Number of pages8
ISBN (Electronic)978-1-6654-3161-3
Publication statusPublished - Feb 2022
MoE publication typeA4 Conference publication
EventIEEE Consumer Communications and Networking Conference - Las Vegas, United States
Duration: 8 Jan 202211 Jan 2022
Conference number: 19


ConferenceIEEE Consumer Communications and Networking Conference
Abbreviated titleCCNC
Country/TerritoryUnited States
CityLas Vegas


  • video streaming
  • deep learning inference
  • object detection


Dive into the research topics of 'Balancing Latency and Accuracy on Deep Video Analytics at the Edge'. Together they form a unique fingerprint.

Cite this