《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (5): 1565-1570.DOI: 10.11772/j.issn.1001-9081.2022040602

• 多媒体计算与计算机仿真 • 上一篇    下一篇

基于状态精细化长短期记忆和注意力机制的社交生成对抗网络用于行人轨迹预测

吴家皋1,2(), 章仕稳1,2, 蒋宇栋1,2, 刘林峰1,2   

  1. 1.南京邮电大学 计算机学院,南京 210023
    2.江苏省大数据安全与智能处理重点实验室(南京邮电大学),南京 210023
  • 收稿日期:2022-04-29 修回日期:2022-07-10 接受日期:2022-07-11 发布日期:2022-08-05 出版日期:2023-05-10
  • 通讯作者: 吴家皋
  • 作者简介:吴家皋(1969—),男,江苏苏州人,副教授,博士,CCF会员,主要研究方向:计算机网络、人工智能 jgwu@njupt.edu.cn
    章仕稳(1996—),男,江苏南京人,硕士研究生,主要研究方向:轨迹预测、深度学习
    蒋宇栋(1999—),男,江苏盐城人,硕士研究生,主要研究方向:轨迹预测、深度学习
    刘林峰(1981—),男,江苏丹阳人,教授,博士,主要研究方向:计算机网络、移动计算。
  • 基金资助:
    国家自然科学基金资助项目(61872191)

Social-interaction GAN for pedestrian trajectory prediction based on state-refinement long short-term memory and attention mechanism

Jiagao WU1,2(), Shiwen ZHANG1,2, Yudong JIANG1,2, Linfeng LIU1,2   

  1. 1.School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing Jiangsu 210023,China
    2.Jiangsu Key Laboratory of Big Data Security and Intelligent Processing (Nanjing University of Posts and Telecommunications),Nanjing Jiangsu 210023,China
  • Received:2022-04-29 Revised:2022-07-10 Accepted:2022-07-11 Online:2022-08-05 Published:2023-05-10
  • Contact: Jiagao WU
  • About author:WU Jiagao, born in 1969, Ph. D., associate professor. His research interests include computer network, artificial intelligence.
    ZHANG Shiwen, born in 1996, M. S. candidate. His research interests include trajectory prediction, deep learning.
    JIANG Yudong, born in 1999, M. S. candidate. His research interests include trajectory prediction, deep learning.
    LIU Linfeng, born in 1981, Ph. D., professor. His research interests include computer network, mobile computing.
  • Supported by:
    National Natural Science Foundation of China(61872191)

摘要:

针对当前行人轨迹预测研究仅考虑影响行人交互因素的问题,基于状态精细化长短期记忆(SR-LSTM)和注意力机制提出一种用于行人轨迹预测的社交生成对抗网络(SRA-SIGAN)模型,利用生成对抗网络(GAN)学习获得目标行人的运动规律。首先,使用SR-LSTM作为位置编码器提取运动意图信息;其次,通过设置速度注意力机制合理地为同一场景中的行人分配影响力,以更好地处理行人的交互;最后,由解码器生成预测的未来轨迹。在多个公开数据集上的测试实验结果表明,SRA-SIGAN模型的总体表现良好。特别是在Zara1数据集上,与SR-LSTM模型相比,SRA-SIGAN模型的平均位移误差(ADE)和最终位移误差(FDE)分别减小了20.0%和10.5%;与社交生成对抗网络(SIGAN)模型相比,SRA-SIGAN的ADE和FDE分别下降了31.7%和24.4%。

关键词: 生成对抗网络, 长短期记忆网络, 行人轨迹预测, 注意力机制, 行人交互

Abstract:

In order to solve the problem of most current research work only considering the factors affecting pedestrian interaction, based on State-Refinement Long Short-Term Memory (SR-LSTM) and attention mechanism, a Social-Interaction Generative Adversarial Network (SIGAN) for pedestrian trajectory prediction was proposed, namely SRA-SIGAN, where GAN was utilized to learn movement patterns of target pedestrians. Firstly, SR-LSTM was used as a location encoder to extract the information of motion intention. Secondly, the influence of pedestrians in the same scene was reasonably assigned by setting the velocity attention mechanism, thereby handling the pedestrian interaction better. Finally, the predicted future trajectory was generated by the decoder. Experimental results on several public datasets show that the performance of SRA-SIGAN model is good on the whole. Specifically on the Zara1 dataset, compared with SR-LSTM model,the Average Displacement Error (ADE)and Final Displacement Error (FDE)of SRA-SIGAN were reduced by 20.0% and 10.5%,respectively;compared with the SIGAN model,the ADE and FDE of SRA-SIGAN were decreased by 31.7% and 24.4%,respectively.

Key words: Generative Adversarial Network (GAN), Long Short-Term Memory (LSTM) network, pedestrian trajectory prediction, attention mechanism, pedestrian interaction

中图分类号: