Tsukada Lab. Computer Network Group

Where Do You Go? Pedestrian Trajectory Prediction using Scene Features

Mohammad Ali Rezaei, Fardin Ayar, Ehsan Javanmardi, Manabu Tsukada, Mahdi Javanmardi

International Conference on Intelligent Computing and its Emerging Applications (ICEA 2024), Tokyo, Japan, pp. 123–128, 2024年11月28日

3行要約

  • 歩行者間の相互作用と周囲の環境の両方から、歩行者の進路を予測する。
  • 疎な歩行者グラフとシーンのセマンティック分割から得た特徴を、交差注意で融合する。
  • 実験で平均位置誤差0.252 m、最終位置誤差0.372 mを得た。

論文の図

図 1: シーン特徴とグラフ特徴を使う歩行者軌跡予測
図 1. シーン特徴とグラフ特徴を使う歩行者軌跡予測出典: Rezaei et al., International Conference on Intelligent Computing and its Emerging Applications (ICEA 2024), 2024(著者版 arXiv:2501.13848) の図 1環境の特徴と歩行者間の関係を交差注意で融合してから、今後の動きを予測することが分かる。

アブストラクト(原文)

Accurate prediction of pedestrian trajectories is crucial for enhancing the safety of autonomous vehicles and reducing traffic fatalities involving pedestrians. While numerous studies have focused on modeling interactions among pedestrians to forecast their movements, the influence of environmental factors and scene-object placements has been comparatively underexplored. In this paper, we present a novel trajectory prediction model that integrates both pedestrian interactions and environmental context to improve prediction accuracy. Our approach captures spatial and temporal interactions among pedestrians within a sparse graph framework. To account for pedestrian-scene interactions, we employ advanced image enhancement and semantic segmentation techniques to extract detailed scene features. These scene and interaction features are then fused through a cross-attention mechanism, enabling the model to prioritize relevant environmental factors that influence pedestrian movements. Finally, a temporal convolutional network processes the fused features to predict future pedestrian trajectories. Experimental results demonstrate that our method significantly outperforms existing state-of-the-art approaches, achieving ADE and FDE values of 0.252 and 0.372 meters, respectively, underscoring the importance of incorporating both social interactions and environmental context in pedestrian trajectory prediction.

引用方法

M. A. Rezaei, F. Ayar, E. Javanmardi, M. Tsukada, and M. Javanmardi, “Where Do You Go? Pedestrian Trajectory Prediction using Scene Features,” in International Conference on Intelligent Computing and its Emerging Applications (ICEA 2024), Tokyo, Japan, 2024, doi: 10.1145/3732437.3732751.

BibTeX

@inproceedings{Rezaei2024SceneFeatures,
  title         = {Where Do You Go? Pedestrian Trajectory Prediction using Scene Features},
  author        = {Mohammad Ali Rezaei and Fardin Ayar and Ehsan Javanmardi and Manabu Tsukada and Mahdi Javanmardi},
  url           = {https://arxiv.org/abs/2501.13848},
  doi           = {10.1145/3732437.3732751},
  eprint        = {2501.13848},
  archiveprefix = {arXiv},
  primaryclass  = {cs.CV},
  year          = {2024},
  date          = {2024-11-28},
  urldate       = {2024-11-28},
  booktitle     = {International Conference on Intelligent Computing and its Emerging Applications (ICEA 2024)},
  pages         = {123--128},
  address       = {Tokyo, Japan},
  abstract      = {Accurate prediction of pedestrian trajectories is crucial for enhancing the safety of autonomous vehicles and reducing traffic fatalities involving pedestrians. While numerous studies have focused on modeling interactions among pedestrians to forecast their movements, the influence of environmental factors and scene-object placements has been comparatively underexplored. In this paper, we present a novel trajectory prediction model that integrates both pedestrian interactions and environmental context to improve prediction accuracy. Our approach captures spatial and temporal interactions among pedestrians within a sparse graph framework. To account for pedestrian-scene interactions, we employ advanced image enhancement and semantic segmentation techniques to extract detailed scene features. These scene and interaction features are then fused through a cross-attention mechanism, enabling the model to prioritize relevant environmental factors that influence pedestrian movements. Finally, a temporal convolutional network processes the fused features to predict future pedestrian trajectories. Experimental results demonstrate that our method significantly outperforms existing state-of-the-art approaches, achieving ADE and FDE values of 0.252 and 0.372 meters, respectively, underscoring the importance of incorporating both social interactions and environmental context in pedestrian trajectory prediction.},
  keywords      = {ITS},
  pubstate      = {published},
  tppubtype     = {inproceedings}
}