To address the issues of frequent identity switches(IDs)and degraded identification accuracy in multi object tracking(MOT)under complex occlusion scenarios,this study proposes an occlusion-robust tracking framework ba...To address the issues of frequent identity switches(IDs)and degraded identification accuracy in multi object tracking(MOT)under complex occlusion scenarios,this study proposes an occlusion-robust tracking framework based on face-pedestrian joint feature modeling.By constructing a joint tracking model centered on“intra-class independent tracking+cross-category dynamic binding”,designing a multi-modal matching metric with spatio-temporal and appearance constraints,and innovatively introducing a cross-category feature mutual verification mechanism and a dual matching strategy,this work effectively resolves performance degradation in traditional single-category tracking methods caused by short-term occlusion,cross-camera tracking,and crowded environments.Experiments on the Chokepoint_Face_Pedestrian_Track test set demonstrate that in complex scenes,the proposed method improves Face-Pedestrian Matching F1 area under the curve(F1 AUC)by approximately 4 to 43 percentage points compared to several traditional methods.The joint tracking model achieves overall performance metrics of IDF1:85.1825%and MOTA:86.5956%,representing improvements of 0.91 and 0.06 percentage points,respectively,over the baseline model.Ablation studies confirm the effectiveness of key modules such as the Intersection over Area(IoA)/Intersection over Union(IoU)joint metric and dynamic threshold adjustment,validating the significant role of the cross-category identity matching mechanism in enhancing tracking stability.Our_model shows a 16.7%frame per second(FPS)drop vs.fairness of detection and re-identification in multiple object tracking(FairMOT),with its cross-category binding module adding aboute 10%overhead,yet maintains near-real-time performance for essential face-pedestrian tracking at small resolutions.展开更多
The intelligent transportation system(ITS)is committed to ensuring safe and effective next-generation traffic throughout a city.However,such efficient operation on urban traffic networks needs the support of big traff...The intelligent transportation system(ITS)is committed to ensuring safe and effective next-generation traffic throughout a city.However,such efficient operation on urban traffic networks needs the support of big traffic data,especially Turning Movement Counts(TMC)at intersections.Generally,TMC data are more challenging to collect due to labor cost and accuracy problems.In this paper,we leverage the capabilities of Unmanned Aerial Vehicles(UAV)to collect real-time TMC data in a cost-efficient way.We proposed a real-time TMC data collection framework based on a live video stream.The vehicle tracking capability is boosted by multiple object tracking based on tracking by detection.In addition,a challenging case study was conducted,and our results demonstrate the feasibility and robustness of the proposed TMC data collection framework.Specifically,with a GTX 1650 graphics card,about 10 FPS can be achieved in real-time for the TMC data collection.The overall accuracy is 91.93%,and the best case is over 98%accurate.In the context of miscounting,the major reason is due to ID switching caused by background occlusion.The proposed framework is expected to provide real-time data for traffic capacity analysis and advanced traffic simulation such as digital twins.展开更多
基金supported by the confidential research grant No.a8317。
文摘To address the issues of frequent identity switches(IDs)and degraded identification accuracy in multi object tracking(MOT)under complex occlusion scenarios,this study proposes an occlusion-robust tracking framework based on face-pedestrian joint feature modeling.By constructing a joint tracking model centered on“intra-class independent tracking+cross-category dynamic binding”,designing a multi-modal matching metric with spatio-temporal and appearance constraints,and innovatively introducing a cross-category feature mutual verification mechanism and a dual matching strategy,this work effectively resolves performance degradation in traditional single-category tracking methods caused by short-term occlusion,cross-camera tracking,and crowded environments.Experiments on the Chokepoint_Face_Pedestrian_Track test set demonstrate that in complex scenes,the proposed method improves Face-Pedestrian Matching F1 area under the curve(F1 AUC)by approximately 4 to 43 percentage points compared to several traditional methods.The joint tracking model achieves overall performance metrics of IDF1:85.1825%and MOTA:86.5956%,representing improvements of 0.91 and 0.06 percentage points,respectively,over the baseline model.Ablation studies confirm the effectiveness of key modules such as the Intersection over Area(IoA)/Intersection over Union(IoU)joint metric and dynamic threshold adjustment,validating the significant role of the cross-category identity matching mechanism in enhancing tracking stability.Our_model shows a 16.7%frame per second(FPS)drop vs.fairness of detection and re-identification in multiple object tracking(FairMOT),with its cross-category binding module adding aboute 10%overhead,yet maintains near-real-time performance for essential face-pedestrian tracking at small resolutions.
基金supported in part by the Research Impact Fund(No.R5007-18)established under the University Grant Committee(UGC)of the Hong Kong Special Administrative Region(HKSAR),Chinasupported in part by the Otto Poon Charitable Foundation Smart Cities Research Institute(Q-CDAS).
文摘The intelligent transportation system(ITS)is committed to ensuring safe and effective next-generation traffic throughout a city.However,such efficient operation on urban traffic networks needs the support of big traffic data,especially Turning Movement Counts(TMC)at intersections.Generally,TMC data are more challenging to collect due to labor cost and accuracy problems.In this paper,we leverage the capabilities of Unmanned Aerial Vehicles(UAV)to collect real-time TMC data in a cost-efficient way.We proposed a real-time TMC data collection framework based on a live video stream.The vehicle tracking capability is boosted by multiple object tracking based on tracking by detection.In addition,a challenging case study was conducted,and our results demonstrate the feasibility and robustness of the proposed TMC data collection framework.Specifically,with a GTX 1650 graphics card,about 10 FPS can be achieved in real-time for the TMC data collection.The overall accuracy is 91.93%,and the best case is over 98%accurate.In the context of miscounting,the major reason is due to ID switching caused by background occlusion.The proposed framework is expected to provide real-time data for traffic capacity analysis and advanced traffic simulation such as digital twins.
