Support vector tracking

被引：744

作者：

Avidan, S ^{[1
]}

机构：

[1] MobilEye Vis Technol LTD, Jerusalem, Israel

来源：

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 2004年 / 26卷 / 08期

关键词：

support vector machines; optic-flow; visual tracking;

D O I：

10.1109/TPAMI.2004.53

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Support Vector Tracking (SVT) integrates the Support Vector Machine (SVM) classifier into an optic-flow-based tracker. Instead of minimizing an intensity difference function between successive frames, SVT maximizes the SVM classification score. To account for large motions between successive frames, we build pyramids from the support vectors and use a coarse-to-fine approach in the classification stage. We show results of using SVT for vehicle tracking in image sequences.

引用

页码：1064 / 1072

页数：9

共 22 条

[1]

ANANDAN P, 1993, MOTION ANAL IMAGE SE, pCH1

[2]

[Anonymous], P 13 INT C MACH LEAR

[3]

BLACK MJ, 1996, P EUR C COMP VIS

[4]

BOTTOU L, 1994, INT C PATT RECOG, P77, DOI 10.1109/ICPR.1994.576879

[5] A tutorial on Support Vector Machines for pattern recognition [J].

Burges, CJC .

DATA MINING AND KNOWLEDGE DISCOVERY, 1998, 2 (02) :121-167

[6]

CORTES C, 1995, MACH LEARN, V20, P273, DOI 10.1023/A:1022627411411

[7]

GONG S, P EUR WORKSH COMB RE

[8]

HEISELE B, 2002, ADV NEURAL INFORMATI, V14

[9]

Joachims T., 1999, P INT C MACH LEARN

[10]

LI SZ, 2002, P 7 EUR C COMP VIS M

← 1 2 3 →