License: Creative Commons Attribution 4.0 International license (CC BY 4.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.ECRTS.2022.2
URN: urn:nbn:de:0030-drops-163199
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2022/16319/
Go to the corresponding LIPIcs Volume Portal

Heo, Seonyeong ; Jeong, Shinnung ; Kim, Hanjun

RTScale: Sensitivity-Aware Adaptive Image Scaling for Real-Time Object Detection

pdf-format:
LIPIcs-ECRTS-2022-2.pdf (2 MB)

Abstract

Real-time object detection is crucial in autonomous driving. To avoid catastrophic accidents, an autonomous car should detect objects with multiple cameras and make decisions within a certain time limit. Object detection systems can meet the real-time constraint by dynamically downsampling input images to proper scales according to their time budget. However, simply applying the same scale to all the images from multiple cameras can cause unnecessary accuracy loss because downsampling can incur a significant accuracy loss for some images.
To reduce the accuracy loss while meeting the real-time constraint, this work proposes RTScale, a new adaptive real-time image scaling scheme that applies different scales to different images reflecting their sensitivities to the scaling and time budget. RTScale infers the sensitivities of multiple images from multiple cameras and determines an appropriate image scale for each image considering the real-time constraint. This work evaluates object detection accuracy and latency with RTScale for two driving datasets. The evaluation results show that RTScale can meet real-time constraints with minimal accuracy loss.

BibTeX - Entry

@InProceedings{heo_et_al:LIPIcs.ECRTS.2022.2,
  author =	{Heo, Seonyeong and Jeong, Shinnung and Kim, Hanjun},
  title =	{{RTScale: Sensitivity-Aware Adaptive Image Scaling for Real-Time Object Detection}},
  booktitle =	{34th Euromicro Conference on Real-Time Systems (ECRTS 2022)},
  pages =	{2:1--2:22},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-239-6},
  ISSN =	{1868-8969},
  year =	{2022},
  volume =	{231},
  editor =	{Maggio, Martina},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2022/16319},
  URN =		{urn:nbn:de:0030-drops-163199},
  doi =		{10.4230/LIPIcs.ECRTS.2022.2},
  annote =	{Keywords: Real-time object detection, Dynamic neural network execution, Adaptive image scaling, Autonomous driving, Self-driving cars}
}

Keywords: Real-time object detection, Dynamic neural network execution, Adaptive image scaling, Autonomous driving, Self-driving cars
Collection: 34th Euromicro Conference on Real-Time Systems (ECRTS 2022)
Issue Date: 2022
Date of publication: 28.06.2022
Supplementary Material: Software (Source Code): https://github.com/seonyheo/darknet

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI