Research Article
Facial Appearance Description Through Facial Landmarks Computation
188 downloads
@INPROCEEDINGS{10.1007/978-3-319-73317-3_13, author={Na Liu and Hao Ge and Lei Song and Huixian Duan}, title={Facial Appearance Description Through Facial Landmarks Computation}, proceedings={Advanced Hybrid Information Processing. First International Conference, ADHIP 2017, Harbin, China, July 17--18, 2017, Proceedings}, proceedings_a={ADHIP}, year={2018}, month={2}, keywords={Facial appearance description Landmarks Facial type}, doi={10.1007/978-3-319-73317-3_13} }
- Na Liu
Hao Ge
Lei Song
Huixian Duan
Year: 2018
Facial Appearance Description Through Facial Landmarks Computation
ADHIP
Springer
DOI: 10.1007/978-3-319-73317-3_13
Abstract
Face appearance descriptions are semantic meaningful characteristics and beneficial for face recognition and retrieval. In this paper, we propose a facial appearance description method which can describe the whole face, chin, eyebrow, eye, nose and mouth type separately. The description is obtained through facial landmarks computation and geometry shape estimation of each part. Based on this method, semantic search of face images can be achieved on face dataset. What’s more, the large scale dataset can be categorized though the facial appearance description before recognition which can help to improve the recognition accuracy and efficiency.
Copyright © 2017–2024 EAI