Anticipating Traffic Accidents with Adaptive Loss and Large-Scale Incident DB
Authors:
- Tomoyuki SuzukiTSTomoyuki Suzuki
- Hirokatsu KataokaHKHirokatsu Kataoka
- Yoshimitsu AokiYAYoshimitsu Aoki
- Yutaka SatohYSYutaka Satoh
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3521-3529
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).Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Agent-Centric Risk Assessment: Accident Anticipation and Risky Region Localization2017
- Efficient Deep Learning for Stereo Matching2016
- Anticipating Visual Representations from Unlabeled Video2016
- Recurrent Neural Networks for driver activity anticipation via sensory-fusion architecture2016
- Fully convolutional networks for semantic segmentation2015
- Anticipating Human Activities Using Object Affordances for Reactive Robotic ResponseIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- The Pascal Visual Object Classes Challenge: A RetrospectiveInternational Journal of Computer Vision, 2014
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014
- Curriculum learning2009