Real-Time Tracking of Fast Moving Weak Object Based on Siamese Network

Sequence	Challenge	Total frames	Weak object frame
Bolt 2	BC、DEF	293	Similar object：1-293
Skiing	IV、SV、OPR、DEF	81	Small object：1-26；32-35；38-81
CarDark	BC、IV	393	Dark object：258-336
MotorRolling	FM、BC、MB、LR	164	Dark object：14-16；40-46；99-105
Matrix	FM、BC、IV、OCC I/OPR、IV、SV	100	Small object：1-57；similar object：1-56 Dark object：3-13；19-30；74-86
Car 1	FM、BC、LR、MBIV、SV	1020	Small object：525-1020；similar object：102-300 Dark object：251-540；614-799；819-897

Table 2. Differences between six algorithms
View table
Table 2. Differences between six algorithms
Algorithms
Feature
Update
Proposed algorithm
Conv3+Conv4+Conv5
Yes
SiamFC^［14］
Conv5
No
KCF^［4］
HOG
No
MEEM^［18］
CIE Lab
Yes
Struck^［19］
Haar
Yes
DLT^［20］
SDAE
Yes

Table 3. Comparison of class algorithms

View table

Table 3. Comparison of class algorithms

Algorithms	OTB100		OTB50		Speed /（frame·s^-1）
Algorithms	AUC	PRE	AUC	PRE	Speed /（frame·s^-1）
Proposed algorithm	0.583	0.783	0.516	0.710	88
SiamFC-MF	0.581	0.774	0.513	0.696	112
SiamFC	0.578	0.770	0.513	0.692	180
CFnet-conv5	0.586	0.711	0.539	0.67	43
MBST	0.599	0.783	0.536	0.718	42.9
Siam-tri	0.590	0.781	0.531	0.713	86.3
SINT++	0.574	0.768	0.624	0.839	1

Tools

Get Citation

Copy Citation Text

Junsong Zheng, Hao Guo, Abiao Li, Jubai An. Real-Time Tracking of Fast Moving Weak Object Based on Siamese Network[J]. Laser & Optoelectronics Progress, 2022, 59(4): 0410011

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites