面向高速公路场景的语义增强跨摄像头车辆跟踪方法
Journal: Engineering Technology Development DOI: 10.32629/etd.v7i2.18958
Abstract
针对高速公路场景下跨摄像头多目标跟踪(MTMCT)中,因光照变化、视角差异及目标遮挡导致的车辆重识别(ReID)精度下降和跨镜关联效率低下的问题,本文提出了一种基于语义增强与向量检索的跨摄像头车辆跟踪方法。首先,在单摄像头视角下,采用YOLOv11目标检测算法结合ByteTrack跟踪器,实现高精度的车辆轨迹提取。其次,为解决跨镜特征匹配的鲁棒性问题,本文改进了CLIP-ReID框架,设计了视觉上下文提示(Visual Context Prompt)模块以融合环境先验信息,并提出语义对齐损失(Semantic Alignment Loss)以缩小视觉特征与文本语义特征之间的异构差距,从而提取出更具鉴别力的车辆外观特征。最后,针对海量车辆特征跨镜匹配带来的计算瓶颈,引入ClickHouse向量数据库进行高维特征的存储与高效检索,完成跨摄像头轨迹的全局关联。在VeRi-776公开数据集与自建高速公路车辆数据集上的实验表明,本文方法在保持较高实时性的同时,显著提升了跨镜多目标跟踪的准确率与鲁棒性。
Keywords
多目标跟踪；车辆重识别；语义增强；向量检索
Full Text
PDF - Viewed/Downloaded: 0 TimesReferences
[1] E.Ristani, F.Solera, R. Zou, R. Cucchiara, and C. Tomasi, "Performance measures and a data set for multi-target,multicamera tracking," in European Conference on Computer Vision (ECCV),Springer,2016,pp.17-35.
[2] A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, "Simple online and realtime tracking," in 2016 IEEE International Conf erence on Image Processing(ICIP),2016,pp.3464-3468.
[3] N.Wojke,A.Bewley,and D.Paulus,"Simple online and realtime tracking with a deep association metric,"in 2017 IEEE Internation al Conference on Image Processing (ICIP),2017,pp.3645-3649.
[4] Y.Zhang,P.Sun, Y. Jiang, D. Yu, F. Weng, Z. Zehuan, Y. Yuan, P.Luo, W. Liu,and X. Wang, "ByteTrack: Multi-object tracking by associating every detection box," in European Conference on Computer Vision (ECCV),Springer,2022,pp.1-21.
[5] X.Liu,W.Liu,T.Mei,and H.Ma,"A deep learning-based appr oach to progressive vehicle re-identification for urban surve illance," in European Conference on Computer Vision (ECCV),Springer,2016,pp.869-884.
[2] A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, "Simple online and realtime tracking," in 2016 IEEE International Conf erence on Image Processing(ICIP),2016,pp.3464-3468.
[3] N.Wojke,A.Bewley,and D.Paulus,"Simple online and realtime tracking with a deep association metric,"in 2017 IEEE Internation al Conference on Image Processing (ICIP),2017,pp.3645-3649.
[4] Y.Zhang,P.Sun, Y. Jiang, D. Yu, F. Weng, Z. Zehuan, Y. Yuan, P.Luo, W. Liu,and X. Wang, "ByteTrack: Multi-object tracking by associating every detection box," in European Conference on Computer Vision (ECCV),Springer,2022,pp.1-21.
[5] X.Liu,W.Liu,T.Mei,and H.Ma,"A deep learning-based appr oach to progressive vehicle re-identification for urban surve illance," in European Conference on Computer Vision (ECCV),Springer,2016,pp.869-884.
Copyright © 2026 郭玮廷
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License