Deo, D and Sen, D (2005) Automatic Recognition of Facial Features and Landmarking of Digital Human Head. In: 6th International Conference on Computer-Aided Industrial Design and Conceptual Design (CAID&CD), MAY 29-JUN 01, 2005, Delft, Netherlands.Full text not available from this repository.
Ergonomic design of products demands accurate human dimensions-anthropometric data. Manual measurement over live subjects, has several limitations like long time, required presence of subjects for every new measurement, physical contact etc. Hence the data currently available is limited and anthropometric data related to facial features is difficult to obtain. In this paper, we discuss a methodology to automatically detect facial features and landmarks from scanned human head models. Segmentation of face into meaningful patches corresponding to facial features is achieved by Watershed algorithms and Mathematical Morphology tools. Many Important physiognomical landmarks are identified heuristically.
|Item Type:||Conference Paper|
|Department/Centre:||Division of Biological Sciences > Molecular Reproduction, Development & Genetics (formed by the merger of DBGL and CRBME)|
|Date Deposited:||08 Mar 2012 09:54|
|Last Modified:||08 Mar 2012 09:54|
Actions (login required)