### Abstract

Shape models are in these days frequently transmitted over Internet and the research of their compression has been started. Considering the large portion of shape model can be normal vectors, a new scheme was recently presented to compress normal vectors using clustering and mixed indexing scheme. Presented in this paper is a mathematical investigation of the scheme to analyze the probability distribution of normal index distances in Normal Index array which is critical for the compression. The probability distribution is formulated so that the values can be easily calculated once the relative probabilities of C, R, E, S, and L op-codes in Edgebreaker are known. It can be shown that the distribution of index distances can be easily transformed into a few measures for the compression performance of the proposed algorithm.

Original language | English |
---|---|

Pages (from-to) | 887-896 |

Number of pages | 10 |

Journal | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |

Volume | 2657 |

State | Published - 2003 Dec 1 |

### Fingerprint

### Cite this

}

*Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)*, vol. 2657, pp. 887-896.

**Probability distribution of index distances in normal index array for normal vector compression.** / Kim, Deok-Soo; Cho, Youngsong; Kim, Donguk; Kim, Hyun.

Research output: Contribution to journal › Article

TY - JOUR

T1 - Probability distribution of index distances in normal index array for normal vector compression

AU - Kim, Deok-Soo

AU - Cho, Youngsong

AU - Kim, Donguk

AU - Kim, Hyun

PY - 2003/12/1

Y1 - 2003/12/1

N2 - Shape models are in these days frequently transmitted over Internet and the research of their compression has been started. Considering the large portion of shape model can be normal vectors, a new scheme was recently presented to compress normal vectors using clustering and mixed indexing scheme. Presented in this paper is a mathematical investigation of the scheme to analyze the probability distribution of normal index distances in Normal Index array which is critical for the compression. The probability distribution is formulated so that the values can be easily calculated once the relative probabilities of C, R, E, S, and L op-codes in Edgebreaker are known. It can be shown that the distribution of index distances can be easily transformed into a few measures for the compression performance of the proposed algorithm.

AB - Shape models are in these days frequently transmitted over Internet and the research of their compression has been started. Considering the large portion of shape model can be normal vectors, a new scheme was recently presented to compress normal vectors using clustering and mixed indexing scheme. Presented in this paper is a mathematical investigation of the scheme to analyze the probability distribution of normal index distances in Normal Index array which is critical for the compression. The probability distribution is formulated so that the values can be easily calculated once the relative probabilities of C, R, E, S, and L op-codes in Edgebreaker are known. It can be shown that the distribution of index distances can be easily transformed into a few measures for the compression performance of the proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=33645337192&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:33645337192

VL - 2657

SP - 887

EP - 896

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -