Real-Time Multiobject Tracking Based on Multiway Concurrency.

Xuan Gong,Hui Wang,Yukun Wu,Zichun Le

doi:10.3390/s21030685

Xuan Gong, Hui Wang + Show 2 more

Open Access

https://doi.org/10.3390/s21030685

Copy DOI

Abstract

This paper explored a pragmatic approach to research the real-time performance of a multiway concurrent multiobject tracking (MOT) system. At present, most research has focused on the tracking of single-image sequences, but in practical applications, multiway video streams need to be processed in parallel by MOT systems. There have been few studies on the real-time performance of multiway concurrent MOT systems. In this paper, we proposed a new MOT framework to solve multiway concurrency scenario based on a tracking-by-detection (TBD) model. The new framework mainly focuses on concurrency and real-time based on limited computing and storage resources, while considering the algorithm performance. For the former, three aspects were studied: (1) Expanded width and depth of tracking-by-detection model. In terms of width, the MOT system can support the process of multiway video sequence at the same time; in terms of depth, image collectors and bounding box collectors were introduced to support batch processing. (2) Considering the real-time performance and multiway concurrency ability, we proposed one kind of real-time MOT algorithm based on directly driven detection. (3) Optimization of system level—we also utilized the inference optimization features of NVIDIA TensorRT to accelerate the deep neural network (DNN) in the tracking algorithm. To trade off the performance of the algorithm, a negative sample (false detection sample) filter was designed to ensure tracking accuracy. Meanwhile, the factors that affect the system real-time performance and concurrency were studied. The experiment results showed that our method has a good performance in processing multiple concurrent real-time video streams.

Highlights

Tracking moving targets involves processing and analyzing the video images captured by photoelectric sensors and making full use of the information to locate and track the target
The multiobject tracking (MOT) with a multicamera is a kind of general vision application
Usually there is a critical demand for real-time tracking performance and system concurrency capability

Summary

Introduction

Tracking moving targets involves processing and analyzing the video (sequence) images captured by photoelectric sensors and making full use of the information to locate and track the target. Target tracking is the foundation of computer vision It has important applications in the field of intelligent monitoring [1], pose estimation [2], motion recognition [3], behavioral analysis [4], automatic driving. In an automatic driving system, target-tracking algorithms need to track cars, pedestrians and animals and predict their positions, speeds, and in the future, perhaps other information. The main task of MOT is to locate multiple targets of interest simultaneously in a given video sequence and maintain their identifications and record their trajectories [5]. MOT needs to solve several key problems: (1) determine the number of targets (usually changing with time) and maintain their respective identification, (2) frequent occlusion, (3) initialization and termination of a track, (4) similar appearance, (5) conflict between multiple objectives [1]

Results

Discussion

Conclusion