ROMA : run-time object detection to maximize real-time accuracy
Abstract
This paper analyzes the effects of dynamically varying video contents and detection latency on the real-time detection accuracy of a detector and proposes a new run-time accuracy variation model, ROMA, based on the findings from the analysis. ROMA is designed to select an optimal detector out of a set of detectors in real time without label information to maximize real-time object detection accuracy. ROMA utilizing four YOLOv4 detectors on an NVIDIA Jetson Nano shows real-time accuracy improvements by 4 to 37% for a scenario of dynamically varying video con-tents and detection latency consisting of MOT17Det and MOT20Det datasets, compared to individual YOLOv4 detectors and two state-of-the-art runtime techniques.
Citation
Lee , J , Varghese , B & Vandierendonck , H 2023 , ROMA : run-time object detection to maximize real-time accuracy . in T Berg & R Farrell (eds) , Proceedings 2023 IEEE/CVF winter conference on applications of computer vision (WACV) : 3-7 January 2023, Waikoloa, Hawaii . , 10030225 , 2023 IEEE/CVF winter conference on applications of computer vision (WACV) , IEEE , Piscataway, NJ , pp. 6394-6403 , IEEE/CVF Winter Conference on Applications of Computer Vision , Waikoloa , Hawaii , United States , 3/01/23 . https://doi.org/10.1109/wacv56688.2023.00634 conference
Publication
Proceedings 2023 IEEE/CVF winter conference on applications of computer vision (WACV)
ISSN
2472-6737Type
Conference item
Description
Funding: This work is supported by the Engineering and Physical Sciences Research Council under grant agreement EP/T022345/1 and by CHIST-ERA under grant agreement CHIST-ERA-18-SDCDN-002 (DiPET).Collections
Items in the St Andrews Research Repository are protected by copyright, with all rights reserved, unless otherwise indicated.