In this work we present a tracking-by-detection system that works under two hard computational constraints: no parallelization and real-time performance. To this aim, we develop a lightweight method for object detection based on background subtraction, window masking and image density projections. The association of detections and trackers is carried out with the Hungarian algorithm on a loss matrix that is constructed according to position, size and appearance. In addition we include heuristics for detecting forming groups as well as incoming and outgoing pedestrians. Finally tracking is performed with particle filters. We test our system on different well-known benchmark datasets. Experimental results reveal that the proposed method is efficient and effective. Specifically, it obtains a processing rate of up to 22 frames per second on average when tracking up to 9 people.
AndrilukaM., RothS. and SchieleB., People-tracking-by-detection and people-detection-by-tracking, in: IEEE Conf on Computer Vision and Pattern Recognition (CVPR), (Jun 2008).
2.
AzabM.M., ShedeedH.A. and HusseinA.S., New technique for online object tracking-by-detection in video, in: IET Image Processing8(12) (2014), 794-803.
3.
BayH., EssA., TuytelaarsT. and Van GoolL., `SURF': Speeded up robust features, in: Computer Vision and Image Understanding (CVIU)110(3) (2008), 346-359.
4.
BernardinK. and StiefelhagenR., Evaluating multiple object tracking performance: The clear MOT metrics, in: J Image Video Process, (Feb 2008), 1-10.
5.
BreitensteinM.D., ReichlinF., LeibeB., Koller-MeierE. and Van GoolL., Online multiperson tracking-by-detection from a single, uncalibrated camera, in: IEEE Transactions on Pattern Analysis and Machine Intelligence33(9) (2011), 1820-1833.
6.
BreitensteinM.D., ReichlinF., LeibeB., Koller-MeierE. and Van GoolL., Robust tracking-by-detection using a detector confidence particle filter, in: IEEE 12th Int Conf on Computer Vision, (Sep 2009), 1515-1522.
7.
ChenX., HenricksonK. and WangY., Kinect-based pedestrian detection for crowded scenes, in: Computer-Aided Civil and Infrastructure Engineering31(3) (2016).
8.
CiarelliP.M., SallesE.O.T. and OliveiraE., Human au-tomatic detection and tracking for outdoor video, in: Inte-grated Computer-Aided Engineering18(4) (2011), 379-390.
9.
ComaniciuD. and MeerP., Mean shift: A robust approach toward feature space analysis, in: IEEE Transactions on Pattern Analysis and Machine Intelligence24(5) (2002), 603-619.
10.
ConchaD., CabidoR., PantrigoJ.J. and MontemayorA.S., Performance evaluation of a 3D multi-view-based particle filter for visual object tracking using GPUs and multicore CPUs, in: Journal of Real-Time Image Processing, in Press, (2014).
11.
CrocianiL. and LämmelG., Finding flows equilibrium in pedestrian environments with a cellular automaton, in: Computer-Aided Civil and Infrastructure Engineering31(6) (2016).
12.
CrowF.C., Summed-area tables for texture mapping, in: Proc of the 11th Annual Conference on Computer Graphics and Interactive Techniques, (1984).
13.
DalalN. and TriggsB., Histograms of oriented gradients for human detection, in: Int Conf on Computer Vision and Pattern Recognition2 (2005), 886-893.
14.
FelzenszwalbP.F., GirshickR.B., McAllersterD. and RamananD., Object detection with discriminatively trained part-based models, in: IEEE Trans on Pattern Analysis and Machine Intelligence32(9) (2010), 1627-1645.
15.
FrancoisJ.B., BerclazJ., FleuretF. and FuaP., Robust people tracking with global trajectory optimization, in: Conf on Computer Vision and Pattern Recognition, (2006), 744-750.
16.
GordonN.J., SalmondD.J. and SmithA.F.M., Novel approach to nonlinear/non-gaussian bayesian state estimation, in: IEEE Proc Radar and Signal Processing140(2) (1993), 107-113.
17.
GuanY., ChenX., YangD. and WuY., Multi-person tracking-by-detection with local particle filtering and global occlusion handling, in: IEEE Int Conf on Multimedia and Expo (ICME), (Jul 2014), 1-6.
18.
HareS. et al., Struck: Structured output tracking with kernels, in: IEEE Trans on Pattern Analysis and Machine Intelligence, in Press, (2015).
19.
HaritaogluI., HarwoodD. and DavisL.S., Real-time surveillance of people and their activities, in: IEEE Trans on Pattern Analysis and Machine Intelligence22(8) (2000), 809-830.
20.
HernándezJ., CabidoR., MontemayorA.S. and PantrigoJ.J., Human activity recognition based on kinematic features, in: Expert Systems (Special Issue: Intelligent Monitoring for People Assistance and Safety)31(4) (2014), 345-353.
21.
HuangC., WuB. and NevatiaR., Robust object tracking by hierarchical association of detection responses, in: ECCV 2008, Part II. LNCS 5303, 788-801.
22.
JalalA.S. and SinghV., The state-of-the-art in visual object, in: Informatica36 (2012), 227-248.
23.
KalmanR.E., A new approach to linear filtering and prediction problems, in: ASME Journal of Basic Engineering, (1960).
24.
KuhnH.W., The hungarian method for the assignment problem, in: Naval Research Logistics Quart2 (1950), 83-87.
25.
LiD., XuL., GoodmanE., XuY. and WuY., Inte-grating a statistical background-foreground extraction algo-rithm and SVM classifier for pedestrian detection and track-ing, in: Integrated Computer-Aided Engineering20(3) (2013), 201-216.
26.
LiP., An efficient particle filter based tracking method using graphics processing unit (GPU), in: J Signal Process Syst68(3) (2012), 317-332.
27.
LimprasertW., WallaceA. and MichaelsonG., Accelerated people tracking using texture in a camera network, in: Int Conf on Computer Vision Theory and Applications, (2012).
28.
OlmedaD., PremebidaC., NunesU., ArmingolJ.M. and de la EscaleraA., Pedestrian detection in far infrared images, in: Integrated Computer-Aided Engineering20(4) (2013), 347-360.
29.
RisticB., ArulampalamS. and GordonN., Beyond the kalman filter: Particle filters for tracking applications, ed., Artech House, 2003.
30.
SanthoshkumarS., KarthikeyanS. and ManjunathB.S., Robust multiple object tracking by detection with interacting markov chain monte carlo, in: 20th IEEE Int Conf on Image Processing (ICIP), (Sep 2013), 2953-2957.
31.
SuzukiS. and AbeK., Topological structural analysis of digitized binary images by border following, in: CVGIP30(1) (1985), 32-46.
32.
TangX. et al., Particle filter track-before-detect implementation on GPU, in: EURASIP Journal on Wireless Communications and Networking1 (Dec 2013), 1-9.
33.
ViolaP. and JonesM., Rapid object detection using a boosted cascade of simple features, in: Proc of Int Conf on Computer Vision and Pattern RecognitionI (2001), 511-518.
34.
WuB. and NevatiaR., Detection and tracking of multiple, partially occluded humans by bayesian combination of edgelet based part detectors, in: Int Journal of Computer Vision (2007).
35.
XiongT., ZhangM., ShuC.W., WongS.C. and ZhangP., High order computational scheme for a dynamic continuum model for bidirectional pedestrian flows, in: Computer-Aided Civil and Infrastructure Engineering26(4) (2011), 297-310.
36.
YilmazA., JavedO. and ShahM., Object tracking: A survey, in: ACM Comput Surv38(4) (2006).
37.
ZhuQ., AvidanS., YehM.C. and ChengK.T., Fast human detection using a cascade of histograms of oriented gradients, in: CVPR (2006), 1491-1498.
