Summary
International Symposium on Nonlinear Theory and Its Applications
2015
Session Number:A2L-E
Session:
Number:A2L-E-4
Motion Vector Estimation of Textureless Objects Exploiting Reaction-Diffusion Cellular Automata
Miho Ushida, Kazuyoshi Ishimura, Alexandre Schmid, Tetsuya Asai, Masato Motomura,
pp.85-88
Publication Date:2015/12/1
Online ISSN:2188-5079
Summary:
Motion vector estimation is applicable to anomaly detection system on security cameras, for instance. It's hard to detect motion vectors for textureless objects because variation of brightness is needed for motion vector estimation. Therefore, we focussed on texture generation on textureless objects by using reaction-diffusion (RD) model. RD model is one of the well-known models of spatial pattern generation. In this study, we simulated motion vector estimation for textureless objects by RD texture generation algorithm that we propose.