Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB

Tomoyuki Suzuki, Hirokatsu Kataoka, Yoshimitsu Aoki, Yutaka Satoh

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

4 Citations (Scopus)

Abstract

In this paper, we propose a novel approach for traffic accident anticipation through (i) Adaptive Loss for Early Anticipation (AdaLEA) and (ii) a large-scale self-annotated incident database for anticipation. The proposed AdaLEA allows a model to gradually learn an earlier anticipation as training progresses. The loss function adaptively assigns penalty weights depending on how early the model can anticipate a traffic accident at each epoch. Additionally, we construct a Near-miss Incident DataBase for anticipation. This database contains an enormous number of traffic near-miss incident videos and annotations for detail evaluation of two tasks, risk anticipation and risk-factor anticipation. In our experimental results, we found our proposal achieved the highest scores for risk anticipation (+6.6% better on mean average precision (mAP) and 2.36 sec earlier than previous work on the average time-to-collision (ATTC)) and risk-factor anticipation (+4.3% better on mAP and 0.70 sec earlier than previous work on ATTC).

Original languageEnglish
Title of host publicationProceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018
PublisherIEEE Computer Society
Pages3521-3529
Number of pages9
ISBN (Electronic)9781538664209
DOIs
Publication statusPublished - 2018 Dec 14
Event31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018 - Salt Lake City, United States
Duration: 2018 Jun 182018 Jun 22

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISSN (Print)1063-6919

Conference

Conference31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018
CountryUnited States
CitySalt Lake City
Period18/6/1818/6/22

Fingerprint

Highway accidents

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Suzuki, T., Kataoka, H., Aoki, Y., & Satoh, Y. (2018). Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB. In Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018 (pp. 3521-3529). [8578469] (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). IEEE Computer Society. https://doi.org/10.1109/CVPR.2018.00371

Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB. / Suzuki, Tomoyuki; Kataoka, Hirokatsu; Aoki, Yoshimitsu; Satoh, Yutaka.

Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018. IEEE Computer Society, 2018. p. 3521-3529 8578469 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

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

Suzuki, T, Kataoka, H, Aoki, Y & Satoh, Y 2018, Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB. in Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018., 8578469, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, pp. 3521-3529, 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018, Salt Lake City, United States, 18/6/18. https://doi.org/10.1109/CVPR.2018.00371
Suzuki T, Kataoka H, Aoki Y, Satoh Y. Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB. In Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018. IEEE Computer Society. 2018. p. 3521-3529. 8578469. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). https://doi.org/10.1109/CVPR.2018.00371
Suzuki, Tomoyuki ; Kataoka, Hirokatsu ; Aoki, Yoshimitsu ; Satoh, Yutaka. / Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB. Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018. IEEE Computer Society, 2018. pp. 3521-3529 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).
@inproceedings{6a385a2fbc104c3693754c1f1023af12,
title = "Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB",
abstract = "In this paper, we propose a novel approach for traffic accident anticipation through (i) Adaptive Loss for Early Anticipation (AdaLEA) and (ii) a large-scale self-annotated incident database for anticipation. The proposed AdaLEA allows a model to gradually learn an earlier anticipation as training progresses. The loss function adaptively assigns penalty weights depending on how early the model can anticipate a traffic accident at each epoch. Additionally, we construct a Near-miss Incident DataBase for anticipation. This database contains an enormous number of traffic near-miss incident videos and annotations for detail evaluation of two tasks, risk anticipation and risk-factor anticipation. In our experimental results, we found our proposal achieved the highest scores for risk anticipation (+6.6{\%} better on mean average precision (mAP) and 2.36 sec earlier than previous work on the average time-to-collision (ATTC)) and risk-factor anticipation (+4.3{\%} better on mAP and 0.70 sec earlier than previous work on ATTC).",
author = "Tomoyuki Suzuki and Hirokatsu Kataoka and Yoshimitsu Aoki and Yutaka Satoh",
year = "2018",
month = "12",
day = "14",
doi = "10.1109/CVPR.2018.00371",
language = "English",
series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
publisher = "IEEE Computer Society",
pages = "3521--3529",
booktitle = "Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018",
address = "United States",

}

TY - GEN

T1 - Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB

AU - Suzuki, Tomoyuki

AU - Kataoka, Hirokatsu

AU - Aoki, Yoshimitsu

AU - Satoh, Yutaka

PY - 2018/12/14

Y1 - 2018/12/14

N2 - In this paper, we propose a novel approach for traffic accident anticipation through (i) Adaptive Loss for Early Anticipation (AdaLEA) and (ii) a large-scale self-annotated incident database for anticipation. The proposed AdaLEA allows a model to gradually learn an earlier anticipation as training progresses. The loss function adaptively assigns penalty weights depending on how early the model can anticipate a traffic accident at each epoch. Additionally, we construct a Near-miss Incident DataBase for anticipation. This database contains an enormous number of traffic near-miss incident videos and annotations for detail evaluation of two tasks, risk anticipation and risk-factor anticipation. In our experimental results, we found our proposal achieved the highest scores for risk anticipation (+6.6% better on mean average precision (mAP) and 2.36 sec earlier than previous work on the average time-to-collision (ATTC)) and risk-factor anticipation (+4.3% better on mAP and 0.70 sec earlier than previous work on ATTC).

AB - In this paper, we propose a novel approach for traffic accident anticipation through (i) Adaptive Loss for Early Anticipation (AdaLEA) and (ii) a large-scale self-annotated incident database for anticipation. The proposed AdaLEA allows a model to gradually learn an earlier anticipation as training progresses. The loss function adaptively assigns penalty weights depending on how early the model can anticipate a traffic accident at each epoch. Additionally, we construct a Near-miss Incident DataBase for anticipation. This database contains an enormous number of traffic near-miss incident videos and annotations for detail evaluation of two tasks, risk anticipation and risk-factor anticipation. In our experimental results, we found our proposal achieved the highest scores for risk anticipation (+6.6% better on mean average precision (mAP) and 2.36 sec earlier than previous work on the average time-to-collision (ATTC)) and risk-factor anticipation (+4.3% better on mAP and 0.70 sec earlier than previous work on ATTC).

UR - http://www.scopus.com/inward/record.url?scp=85062847710&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062847710&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2018.00371

DO - 10.1109/CVPR.2018.00371

M3 - Conference contribution

AN - SCOPUS:85062847710

T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

SP - 3521

EP - 3529

BT - Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2018

PB - IEEE Computer Society

ER -