Automatic Extraction of Frontal Facial Features

Automatic Extraction of Frontal Facial Features

Cached

Download Links

by Vladimir Vezhnevets , Stanislav Soldatov , Anna Degtiareva , In Kyu Park

Citations:

1 - 0 self

Versions

Version History

Metadata Version 2

User correction supplied by SystemCorrections

Datum	Value	Source
TITLE	Automatic Extraction of Frontal Facial Features	user correction - Legacy Corrections
AUTHOR NAME	Vladimir Vezhnevets	user correction - Legacy Corrections
AUTHOR AFFIL	Dept. of Computational Mathematics and Cybernetics; Moscow State University	user correction - Legacy Corrections
AUTHOR ADDR	Moscow 119899, RUSSIAN FEDERATION	user correction - Legacy Corrections
AUTHOR NAME	Stanislav Soldatov	user correction - Legacy Corrections
AUTHOR AFFIL	Dept. of Computational Mathematics and Cybernetics; Moscow State University	user correction - Legacy Corrections
AUTHOR ADDR	Moscow 119899, RUSSIAN FEDERATION	user correction - Legacy Corrections
AUTHOR NAME	Anna Degtiareva	user correction - Legacy Corrections
AUTHOR AFFIL	Dept. of Computational Mathematics and Cybernetics; Moscow State University	user correction - Legacy Corrections
AUTHOR ADDR	Moscow 119899, RUSSIAN FEDERATION	user correction - Legacy Corrections
AUTHOR NAME	In Kyu Park	user correction - Legacy Corrections
ABSTRACT	This paper describes algorithms for face and facial features detection in still frontal images. It is designed for the task of automatic image-based 3-D face modelling. This requires the detection to be accurate enough to produce exact facial features and robust to images of widely varying quality and picture taking conditions. The facial feature detection task is solved in several steps. First, facial area is detected using a novel method based on skin color segmentation and adaptive ellipse fitting. Next, eye positions are estimated by finding eye-shaped and eye-sized areas of red channel sharp changes. Finally, exact facial contours of eyes, eyebrows, nose, mouth, chin, and cheek are estimated by employing deformable models, template matching, and color segmentation. The main contribution of this paper is a set of innovations in face and facial feature detection algorithms that achieve high detection robustness and accuracy.	user correction - Legacy Corrections
CITATIONS	12 found	ParsCit 1.0