《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (5): 1480-1487.DOI: 10.11772/j.issn.1001-9081.2024050650
• 人工智能 • 上一篇
收稿日期:
2024-05-23
修回日期:
2024-08-28
接受日期:
2024-09-05
发布日期:
2024-09-13
出版日期:
2025-05-10
通讯作者:
杨小军
作者简介:
陈满(2000—),男,河南南阳人,硕士研究生,CCF会员,主要研究方向:轨迹预测、人工智能基金资助:
Man CHEN, Xiaojun YANG(), Huimin YANG
Received:
2024-05-23
Revised:
2024-08-28
Accepted:
2024-09-05
Online:
2024-09-13
Published:
2025-05-10
Contact:
Xiaojun YANG
About author:
CHEN Man, born in 2000, M. S. candidate. His research interests include trajectory prediction, artificial intelligence.Supported by:
摘要:
针对行人轨迹预测研究中仅关注历史轨迹的交互信息,而忽略了终点交互信息的问题,提出一种基于图卷积网络(GCN)和终点诱导(Endpoint Induction)的行人轨迹预测模型GCN-EI。首先,在训练集上使用分类方法学习行人未来可能的加权终点分布;其次,将可能的终点与它们对应的历史轨迹相连接,并使用基于注意力机制和终点条件的GCN在更长的时间跨度上提取行人的交互特征,同时使用个体特征模块提取行人的内在运动特征;最后通过时间内推卷积预测行人的未来轨迹。在ETH和UCY数据集上对模型进行的测试结果表明,相较于STITD-GCN(Spatio-Temporal Interaction and Trajectory Distribution GCN)模型,所提模型在平均位移误差(ADE)和最终位移误差(FDE)上分别下降了4.5%和5.0%;相较于采用分类方法的PCCSNet(Prediction via modality Clustering, Classification and Synthesis Network)模型,在FDE上下降了9.5%。
中图分类号:
陈满, 杨小军, 杨慧敏. 基于图卷积网络和终点诱导的行人轨迹预测[J]. 计算机应用, 2025, 45(5): 1480-1487.
Man CHEN, Xiaojun YANG, Huimin YANG. Pedestrian trajectory prediction based on graph convolutional network and endpoint induction[J]. Journal of Computer Applications, 2025, 45(5): 1480-1487.
模型 | ETH | HOTEL | UNIV | ZARA1 | ZARA2 | 平均 | ||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
ADE | FDE | ADE | FDE | ADE | FDE | ADE | FDE | ADE | FDE | ADE | FDE | |
Social-STGCNN | 0.64 | 1.11 | 0.49 | 0.85 | 0.44 | 0.79 | 0.34 | 0.53 | 0.30 | 0.48 | 0.44 | 0.75 |
PECNet | 0.54 | 0.87 | 0.18 | 0.24 | 0.35 | 0.60 | 0.22 | 0.39 | 0.17 | 0.30 | 0.29 | 0.48 |
SGCN | 0.63 | 1.03 | 0.32 | 0.50 | 0.37 | 0.70 | 0.29 | 0.53 | 0.25 | 0.45 | 0.37 | 0.65 |
PCCSNet | 0.28 | 0.54 | 0.11 | 0.19 | 0.29 | 0.60 | 0.21 | 0.44 | 0.15 | 0.34 | 0.21 | 0.42 |
VDRGCN | 0.62 | 0.81 | 0.27 | 0.37 | 0.38 | 0.58 | 0.29 | 0.42 | 0.21 | 0.32 | 0.35 | 0.50 |
CTSGI | 0.30 | 0.57 | 0.11 | 0.20 | 0.25 | 0.54 | 0.22 | 0.49 | 0.17 | 0.39 | 0.21 | 0.44 |
STITD-GCN | 0.30 | 0.52 | 0.18 | 0.32 | 0.28 | 0.52 | 0.19 | 0.34 | 0.15 | 0.28 | 0.22 | 0.40 |
GCN-EI | 0.37 | 0.56 | 0.12 | 0.19 | 0.25 | 0.44 | 0.19 | 0.40 | 0.13 | 0.29 | 0.21 | 0.38 |
表1 ETH与UCY数据集上本文模型与各种基线模型的性能对比 ( m)
Tab. 1 Performance comparison of proposed model and various baseline models on ETH and UCY datasets
模型 | ETH | HOTEL | UNIV | ZARA1 | ZARA2 | 平均 | ||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
ADE | FDE | ADE | FDE | ADE | FDE | ADE | FDE | ADE | FDE | ADE | FDE | |
Social-STGCNN | 0.64 | 1.11 | 0.49 | 0.85 | 0.44 | 0.79 | 0.34 | 0.53 | 0.30 | 0.48 | 0.44 | 0.75 |
PECNet | 0.54 | 0.87 | 0.18 | 0.24 | 0.35 | 0.60 | 0.22 | 0.39 | 0.17 | 0.30 | 0.29 | 0.48 |
SGCN | 0.63 | 1.03 | 0.32 | 0.50 | 0.37 | 0.70 | 0.29 | 0.53 | 0.25 | 0.45 | 0.37 | 0.65 |
PCCSNet | 0.28 | 0.54 | 0.11 | 0.19 | 0.29 | 0.60 | 0.21 | 0.44 | 0.15 | 0.34 | 0.21 | 0.42 |
VDRGCN | 0.62 | 0.81 | 0.27 | 0.37 | 0.38 | 0.58 | 0.29 | 0.42 | 0.21 | 0.32 | 0.35 | 0.50 |
CTSGI | 0.30 | 0.57 | 0.11 | 0.20 | 0.25 | 0.54 | 0.22 | 0.49 | 0.17 | 0.39 | 0.21 | 0.44 |
STITD-GCN | 0.30 | 0.52 | 0.18 | 0.32 | 0.28 | 0.52 | 0.19 | 0.34 | 0.15 | 0.28 | 0.22 | 0.40 |
GCN-EI | 0.37 | 0.56 | 0.12 | 0.19 | 0.25 | 0.44 | 0.19 | 0.40 | 0.13 | 0.29 | 0.21 | 0.38 |
数据集 | 不同聚类算法的FDE/m | |||
---|---|---|---|---|
小批量K-Means | 层次聚类 | BIRCH | 高斯混合模型 | |
ETH | 0.56 | 0.54 | 0.68 | 0.57 |
HOTEL | 0.19 | 0.23 | 0.24 | 0.19 |
UNIV | 0.44 | 0.44 | 0.46 | 0.43 |
ZARA1 | 0.41 | 0.43 | 0.45 | 0.42 |
ZARA2 | 0.30 | 0.31 | 0.35 | 0.30 |
平均 | 0.38 | 0.39 | 0.44 | 0.38 |
Time/min | 1.50 | 22.00 | 0.10 | 145.20 |
表2 不同聚类算法对性能的影响
Tab. 2 Influence of different clustering algorithms on performance
数据集 | 不同聚类算法的FDE/m | |||
---|---|---|---|---|
小批量K-Means | 层次聚类 | BIRCH | 高斯混合模型 | |
ETH | 0.56 | 0.54 | 0.68 | 0.57 |
HOTEL | 0.19 | 0.23 | 0.24 | 0.19 |
UNIV | 0.44 | 0.44 | 0.46 | 0.43 |
ZARA1 | 0.41 | 0.43 | 0.45 | 0.42 |
ZARA2 | 0.30 | 0.31 | 0.35 | 0.30 |
平均 | 0.38 | 0.39 | 0.44 | 0.38 |
Time/min | 1.50 | 22.00 | 0.10 | 145.20 |
聚类簇数 | 不同数据集的FDE/m | 平均FDE/m | ||||
---|---|---|---|---|---|---|
ETH | HOTEL | UNIV | ZARA1 | ZARA2 | ||
350 | 0.59 | 0.23 | 0.46 | 0.44 | 0.31 | 0.41 |
400 | 0.56 | 0.21 | 0.45 | 0.41 | 0.30 | 0.39 |
450 | 0.56 | 0.19 | 0.44 | 0.41 | 0.30 | 0.38 |
500 | 0.59 | 0.21 | 0.47 | 0.42 | 0.32 | 0.40 |
550 | 0.59 | 0.22 | 0.48 | 0.45 | 0.36 | 0.42 |
表3 不同聚类簇数对FDE的影响
Tab. 3 Influence of different cluster numbers on FDE
聚类簇数 | 不同数据集的FDE/m | 平均FDE/m | ||||
---|---|---|---|---|---|---|
ETH | HOTEL | UNIV | ZARA1 | ZARA2 | ||
350 | 0.59 | 0.23 | 0.46 | 0.44 | 0.31 | 0.41 |
400 | 0.56 | 0.21 | 0.45 | 0.41 | 0.30 | 0.39 |
450 | 0.56 | 0.19 | 0.44 | 0.41 | 0.30 | 0.38 |
500 | 0.59 | 0.21 | 0.47 | 0.42 | 0.32 | 0.40 |
550 | 0.59 | 0.22 | 0.48 | 0.45 | 0.36 | 0.42 |
变体 | ADE | 平均 | ||||
---|---|---|---|---|---|---|
ETH | HOTEL | UNIV | ZARA1 | ZARA2 | ||
a | 0.91 | 0.56 | 0.64 | 0.42 | 0.36 | 0.58 |
b | 0.41 | 0.13 | 0.25 | 0.20 | 0.15 | 0.23 |
c | 0.45 | 0.12 | 0.28 | 0.23 | 0.18 | 0.25 |
d | 0.44 | 0.12 | 0.26 | 0.19 | 0.15 | 0.23 |
GCN-EI | 0.37 | 0.12 | 0.25 | 0.19 | 0.13 | 0.21 |
表4 GCN-EI不同变体的平均ADE对比 ( m)
Tab. 4 Comparison of average ADE among different variants of GCN-EI
变体 | ADE | 平均 | ||||
---|---|---|---|---|---|---|
ETH | HOTEL | UNIV | ZARA1 | ZARA2 | ||
a | 0.91 | 0.56 | 0.64 | 0.42 | 0.36 | 0.58 |
b | 0.41 | 0.13 | 0.25 | 0.20 | 0.15 | 0.23 |
c | 0.45 | 0.12 | 0.28 | 0.23 | 0.18 | 0.25 |
d | 0.44 | 0.12 | 0.26 | 0.19 | 0.15 | 0.23 |
GCN-EI | 0.37 | 0.12 | 0.25 | 0.19 | 0.13 | 0.21 |
变体 | FDE | 平均 | ||||
---|---|---|---|---|---|---|
ETH | HOTEL | UNIV | ZARA1 | ZARA2 | ||
a | 1.90 | 1.06 | 1.31 | 0.89 | 0.79 | 1.19 |
GCN-EI | 0.56 | 0.19 | 0.44 | 0.40 | 0.29 | 0.38 |
表5 GCN-EI不同变体的平均FDE对比 ( m)
Tab. 5 Comparison of average FDE among different variants of GCN-EI
变体 | FDE | 平均 | ||||
---|---|---|---|---|---|---|
ETH | HOTEL | UNIV | ZARA1 | ZARA2 | ||
a | 1.90 | 1.06 | 1.31 | 0.89 | 0.79 | 1.19 |
GCN-EI | 0.56 | 0.19 | 0.44 | 0.40 | 0.29 | 0.38 |
1 | LIANG J, JIANG L, NIEBLES J C, et al. Peeking into the future: predicting future person activities and locations in videos[C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Piscataway: IEEE, 2019: 5718-5727. |
2 | YU K, LIN L, ALAZAB M, et al. Deep learning-based traffic safety solution for a mixture of autonomous and manual vehicles in a 5G-enabled intelligent transportation system[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(7): 4337-4347. |
3 | BASTANI V, MARCENARO L, REGAZZONI C S. Online nonparametric Bayesian activity mining and analysis from surveillance video[J]. IEEE Transactions on Image Processing, 2016, 25(5): 2089-2102. |
4 | CHEN X, LUO F, ZHAO F, et al. Goal-guided and interaction-aware state refinement graph attention network for multi-agent trajectory prediction[J]. IEEE Robotics and Automation Letters, 2024, 9(1): 57-64. |
5 | CHEN W, YANG Z, XUE L, et al. Multimodal pedestrian trajectory prediction using probabilistic proposal network[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2023, 33(6): 2877-2891. |
6 | MANGALAM K, AN Y, GIRASE H, et al. From goals, waypoints & paths to long term human trajectory forecasting[C]// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2021: 15213-15222. |
7 | LI J, MA H, TOMIZUKA M. Conditional generative neural system for probabilistic trajectory prediction[C]// Proceedings of the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway: IEEE, 2019: 6150-6156. |
8 | SOHN K, YAN X, LEE H. Learning structured output representation using deep conditional generative models[C]// Proceedings of the 29th International Conference on Neural Information Processing Systems — Volume 2. Cambridge: MIT Press, 2015: 3483-3491. |
9 | LEE N, CHOI W, VERNAZA P, et al. DESIRE: distant future prediction in dynamic scenes with interacting agents[C]// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 2165-2174. |
10 | LI Y, LIANG R, WEI W, et al. Temporal pyramid network with spatial-temporal attention for pedestrian trajectory prediction[J]. IEEE Transactions on Network Science and Engineering, 2022, 9(3): 1006-1019. |
11 | GUPTA A, JOHNSON J, LI F F, et al. Social GAN: socially acceptable trajectories with generative adversarial networks[C]// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 2255-2264. |
12 | GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]// Proceedings of the 28th International Conference on Neural Information Processing Systems — Volume 2. Cambridge: MIT Press, 2014: 2672-2680. |
13 | GIULIARI F, HASAN I, CRISTANI M, et al. Transformer networks for trajectory forecasting[C]// Proceedings of the 25th International Conference on Pattern Recognition. Piscataway: IEEE, 2020: 10335-10342. |
14 | PHAN-MINH T, GRIGORE E C, BOULTON F A, et al. CoverNet: multimodal behavior prediction using trajectory sets[C]// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2020: 14062-14071. |
15 | ZHAO H, GAO J, LAN T, et al. TNT: target-driven trajectory prediction[C]// Proceedings of the 4th Conference on Robot Learning. New York: JMLR.org, 2021: 895-904. |
16 | SUN J, LI Y, FANG H S, et al. Three steps to multimodal trajectory prediction: modality clustering, classification and synthesis[C]// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2021: 13250-13259. |
17 | FANG L, JIANG Q, SHI J, et al. TPNet: trajectory proposal network for motion prediction[C]// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2020: 6796-6805. |
18 | LIAN J, YU F, LI L, et al. Causal temporal-spatial pedestrian trajectory prediction with goal point estimation and contextual interaction[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(12): 24499-24509. |
19 | KIPF T N, WELLING M. Semi-supervised classification with graph convolutional networks[EB/OL]. [2024-06-20].. |
20 | PELLEGRINI S, ESS A, SCHINDLER K, et al. You'll never walk alone: modeling social behavior for multi-target tracking[C]// Proceedings of the IEEE 12th International Conference on Computer Vision. Piscataway: IEEE, 2009: 261-268. |
21 | LERNER A, CHRYSANTHOU Y, LISCHINSKI D. Crowds by example[J]. Computer Graphics Forum, 2007, 26(3): 655-664. |
22 | MANGALAM K, GIRASE H, AGARWAL S, et al. It is not the journey but the destination: endpoint conditioned trajectory prediction[C]// Proceedings of the 2020 European Conference on Computer Vision LNCS 12347. Cham: Springer, 2020: 759-776. |
23 | VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook: Curran Associates Inc., 2017: 6000-6010. |
24 | MOHAMED A, QIAN K, ELHOSEINY M, et al. Social-STGCNN: a social spatio-temporal graph convolutional neural network for human trajectory prediction[C]// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2020: 14412-14420. |
25 | SHI L, WANG L, LONG C, et al. SGCN: sparse graph convolution network for pedestrian trajectory prediction[C]// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2021: 8990-8999. |
26 | MacQUEEN J. Some methods for classification and analysis of multivariate observations[C]// Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics. Berkeley: University of California Press, 1967: 281-297. |
27 | ALAHI A, GOEL K, RAMANATHAN V, et al. Social LSTM: human trajectory prediction in crowded spaces[C]// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 961-971. |
28 | ZHANG P, OUYANG W, ZHANG P, et al. SR-LSTM: state refinement for LSTM towards pedestrian trajectory prediction[C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 12077-12086. |
29 | SU Y, DU J, LI Y, et al. Trajectory forecasting based on prior-aware directed graph convolutional neural network[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(9): 16773-16785. |
30 | WANG R, SONG X, HU Z, et al. Spatio-temporal interaction aware and trajectory distribution aware graph convolution network for pedestrian multimodal trajectory prediction[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: No.5001211. |
[1] | 王泉, 陆啟想, 施珮. 用于交通流量预测的多图扩散注意力网络[J]. 《计算机应用》唯一官方网站, 2025, 45(5): 1472-1479. |
[2] | 李慧, 贾炳志, 王晨曦, 董子宇, 李纪龙, 仲兆满, 陈艳艳. 基于Swin Transformer的生成对抗网络水下图像增强模型[J]. 《计算机应用》唯一官方网站, 2025, 45(5): 1439-1446. |
[3] | 王利琴, 耿智雷, 李英双, 董永峰, 边萌. 基于路径和增强三元组文本的开放世界知识推理模型[J]. 《计算机应用》唯一官方网站, 2025, 45(4): 1177-1183. |
[4] | 党伟超, 宋楚君, 高改梅, 刘春霞. 基于级联残差图卷积网络的多行为推荐[J]. 《计算机应用》唯一官方网站, 2025, 45(4): 1223-1231. |
[5] | 郭诗月, 党建武, 王阳萍, 雍玖. 结合注意力机制和多尺度特征融合的三维手部姿态估计[J]. 《计算机应用》唯一官方网站, 2025, 45(4): 1293-1299. |
[6] | 胡婕, 郑启扬, 孙军, 张龑. 基于多标签关系图和局部动态重构学习的多标签分类模型[J]. 《计算机应用》唯一官方网站, 2025, 45(4): 1104-1112. |
[7] | 徐春, 吉双焱, 马欢, 孙恩威, 王萌萌, 苏明钰. 基于知识图谱和对话结构的问诊推荐方法[J]. 《计算机应用》唯一官方网站, 2025, 45(4): 1157-1168. |
[8] | 张李伟, 梁泉, 胡禹涛, 朱乔乐. 基于分组卷积的通道重洗注意力机制[J]. 《计算机应用》唯一官方网站, 2025, 45(4): 1069-1076. |
[9] | 姜坤元, 李小霞, 王利, 曹耀丹, 张晓强, 丁楠, 周颖玥. 引入解耦残差自注意力的边界交叉监督语义分割网络[J]. 《计算机应用》唯一官方网站, 2025, 45(4): 1120-1129. |
[10] | 耿海军, 董赟, 胡治国, 池浩田, 杨静, 尹霞. 基于Attention-1DCNN-CE的加密流量分类方法[J]. 《计算机应用》唯一官方网站, 2025, 45(3): 872-882. |
[11] | 张天骐, 谭霜, 沈夕文, 唐娟. 融合注意力机制和多尺度特征的图像水印方法[J]. 《计算机应用》唯一官方网站, 2025, 45(2): 616-623. |
[12] | 富坤, 应世聪, 郑婷婷, 屈佳捷, 崔静远, 李建伟. 面向小样本节点分类的图数据增强方法[J]. 《计算机应用》唯一官方网站, 2025, 45(2): 392-402. |
[13] | 蔡启健, 谭伟. 语义图增强的多模态推荐算法[J]. 《计算机应用》唯一官方网站, 2025, 45(2): 421-427. |
[14] | 李严, 叶冠华, 李雅文, 梁美玉. 基于丰度协调技术的企业ESG指标预测模型[J]. 《计算机应用》唯一官方网站, 2025, 45(2): 670-676. |
[15] | 王地欣, 王佳昊, 李敏, 陈浩, 胡光耀, 龚宇. 面向水声通信网络的异常攻击检测[J]. 《计算机应用》唯一官方网站, 2025, 45(2): 526-533. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||