Research Article

Neuromorphic Vision Based Multivehicle Detection and Tracking for Intelligent Transportation System

Table 4

The tracking performance for EventSeq-Vehicle1 using different tracking methods (SORT, GM-PHD, GM-CPHD, and PDAF) with detection through DBSCAN clustering. The impact of time intervals from 10ms to 30ms was studied.

Tis TrackerMOTAMOTPMTPTMLFP FNIDsFM

10msSORT36.2%69.2%879202891163691461302
10msGM-PHD24.0%69.1%1852149241664915414097
10msGM-CPHD21.1%69.2%389157480156219003616
10msPDAF20.9%69.1%086215653182281584678

20msSORT35.0%70.2%1871182905689392444
20msGM-PHD35.1%70.6%18701925237019323770
20msGM-CPHD25.7%70.5%12752039747180152716
20msPDAF24.5%70.4%4802335767815951371

30msSORT28.5%70.4%1269261950519094265
30msGM-PHD23.6%70.8%14672623235224190478
30msGM-CPHD18.3%70.7%8762328705259135481
30msPDAF19.3%70.5%176302402570166900