Join Our Newsletter

Free Online Productivity Tools
i2Speak
i2Symbol
i2OCR
iTex2Img
iWeb2Print
iWeb2Shot
i2Type
iPdf2Split
iPdf2Merge
i2Bopomofo
i2Arabic
i2Style
i2Image
i2PDF
iLatex2Rtf
Sci2ools

ACCV

2007

Springer

2007

Springer

Abstract. The trace quotient problem arises in many applications in pattern classiﬁcation and computer vision, e.g., manifold learning, low-dimension embedding, etc. The task is to solve a optimization problem involving maximizing the ratio of two traces, i.e., maxW Tr(f(W ))/Tr(h(W )). This optimization problem itself is non-convex in general, hence it is hard to solve it directly. Conventionally, the trace quotient objective function is replaced by a much simpler quotient trace formula, i.e., maxW Tr h(W )−1 f(W ) ¡ , which accommodates a much simpler solution. However, the result is no longer optimal for the original problem setting, and some desirable properties of the original problem are lost. In this paper we proposed a new formulation for solving the trace quotient problem directly. We reformulate the original non-convex problem such that it can be solved by efﬁciently solving a sequence of semideﬁnite feasibility problems. The solution is therefore globally optimal...

Related Content

Added |
18 Oct 2010 |

Updated |
18 Oct 2010 |

Type |
Conference |

Year |
2007 |

Where |
ACCV |

Authors |
Chunhua Shen, Hongdong Li, Michael J. Brooks |

Comments (0)