Presentation 2020-03-10
EM Algorithm for Mixture Models in Shape Analysis
Kazunori Iwata,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The mixture model is a well-known probabilistic model in statistical modeling to express a population that is composed of subpopulations. The EM algorithm plays a vital role in the estimation for the set of parameters in the mixture model. Procrustes analysis is a type of shape analysis that is not affected by several of the similarity transformations to shapes. In Procrustes analysis, the ordinary Procrustes sum of squares can be used to measure the distance between shapes. Thus, in this paper, our aim is to present a novel shape clustering method as a type of Procrustes analysis by incorporating it into an individual component of the mixture model and by deriving an EM algorithm for the model. Through several experiments using datasets of line drawings and image outlines, we demonstrate that our shape clustering works well compared with that based on a typical vector-based distance measure that is affected by similarity transformations.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Shape clustering / Procrustes analysis / EM algorithm
Paper # IBISML2019-33
Date of Issue 2020-03-03 (IBISML)

Conference Information
Committee IBISML
Conference Date 2020/3/10(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Kyoto University
Topics (in Japanese) (See Japanese page)
Topics (in English) Machine learning, etc.
Chair Hisashi Kashima(Kyoto Univ.)
Vice Chair Masashi Sugiyama(Univ. of Tokyo) / Koji Tsuda(Univ. of Tokyo)
Secretary Masashi Sugiyama(Nagoya Inst. of Tech.) / Koji Tsuda(AIST)
Assistant Tomoharu Iwata(NTT) / Shigeyuki Oba(Kyoto Univ.)

Paper Information
Registration To Technical Committee on Infomation-Based Induction Sciences and Machine Learning
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) EM Algorithm for Mixture Models in Shape Analysis
Sub Title (in English)
Keyword(1) Shape clustering
Keyword(2) Procrustes analysis
Keyword(3) EM algorithm
1st Author's Name Kazunori Iwata
1st Author's Affiliation Hiroshima City University(Hiroshima City Univ.)
Date 2020-03-10
Paper # IBISML2019-33
Volume (vol) vol.119
Number (no) IBISML-476
Page pp.pp.1-7(IBISML),
#Pages 7
Date of Issue 2020-03-03 (IBISML)