Identifying and Extracting Pedestrian Behavior in Critical Traffic Situations

📄 arXiv: 2402.02533v1 📥 PDF

作者: Martin Schachner, Bernd Schneider, Fabian Weissenbacher, Nadezda Kirillova, Horst Possegger, Horst Bischof, Corina Klug

分类: cs.RO

发布日期: 2024-02-04

备注: 7 pages, 8 figures, ITSC 2023 accepted


💡 一句话要点

提出一种方法以识别和提取关键交通情况下的行人行为

🎯 匹配领域: 支柱七:动作重定向 (Motion Retargeting)

关键词: 行人行为识别 交通安全 视频分析 关键场景提取 自动驾驶 智能交通系统

📋 核心要点

  1. 现有方法在识别关键交通情况下的行人行为时,缺乏有效的实时分析手段,导致安全隐患未能及时识别。
  2. 论文提出通过结合后侵入时间和运动适应度量,自动分析行人和车辆的互动,从而识别关键场景。
  3. 实验结果表明,应用新方法后,仅识别出21个关键场景,显著减少了误报,提升了关键互动的识别精度。

📝 摘要(中文)

更好地理解关键交通情况下的行人互动行为对于提升行人安全系统至关重要。本文提出了一种从基于摄像头的观察系统中提取重要行人-车辆互动的方法。通过分析110小时的视频记录,结合后侵入时间和新引入的运动适应度量,识别出259个潜在场景。在95%的情况下,未观察到行人行为的适应,但应用运动适应度量后,仅剩21个关键场景,其中7个显示了行人与车辆的关键互动。这些结果为行人行为模型的开发提供了数据支持。

🔬 方法详解

问题定义:本文旨在解决如何有效识别和提取关键交通情况下行人行为的问题。现有方法在实时分析和准确识别方面存在不足,无法全面捕捉行人与车辆的互动。

核心思路:论文的核心思路是结合后侵入时间与运动适应度量,通过自动分析行人轨迹,建立时间和空间关系,从而识别出关键的行人-车辆互动场景。

技术框架:整体架构包括视频数据采集、轨迹重建、时间与空间关系分析、关键场景识别等主要模块。首先,通过摄像头获取交通视频,然后自动提取行人轨迹,最后应用新引入的度量进行关键场景的筛选。

关键创新:最重要的技术创新在于引入了运动适应度量,与传统的后侵入时间结合,能够更准确地识别出行人对车辆的反应,尤其是在关键情况下的行为变化。

关键设计:在参数设置上,后侵入时间阈值设定为2秒,运动适应度量的具体计算方法也进行了详细设计,以确保能够有效区分关键与非关键场景。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示,应用新方法后,仅识别出21个关键场景,相较于初步筛选的259个场景,减少了约91%的误报率。同时,手动验证发现7个场景确实存在关键行人-车辆互动,表明该方法在实际应用中的有效性。

🎯 应用场景

该研究的潜在应用领域包括智能交通系统、自动驾驶车辆的行人检测与预测、以及城市交通安全管理等。通过更准确地识别行人行为,能够为交通安全系统提供更可靠的数据支持,进而提升行人安全性,减少交通事故发生率。

📄 摘要(原文)

A better understanding of interactive pedestrian behavior in critical traffic situations is essential for the development of enhanced pedestrian safety systems. Real-world traffic observations play a decisive role in this, since they represent behavior in an unbiased way. In this work, we present an approach of how a subset of very considerable pedestrian-vehicle interactions can be derived from a camera-based observation system. For this purpose, we have examined road user trajectories automatically for establishing temporal and spatial relationships, using 110h hours of video recordings. In order to identify critical interactions, our approach combines the metric post-encroachment time with a newly introduced motion adaption metric. From more than 11,000 reconstructed pedestrian trajectories, 259 potential scenarios remained, using a post-encroachment time threshold of 2s. However, in 95% of cases, no adaptation of the pedestrian behavior was observed due to avoiding criticality. Applying the proposed motion adaption metric, only 21 critical scenarios remained. Manual investigations revealed that critical pedestrian vehicle interactions were present in 7 of those. They were further analyzed and made publicly available for developing pedestrian behavior models3. The results indicate that critical interactions in which the pedestrian perceives and reacts to the vehicle at a relatively late stage can be extracted using the proposed method.