Component Localization in Face Alignment
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2011, Vol 0, Issue 1
Abstract
- Face alignment is a significant problem in the processing of face image, and Active Shape Model (ASM) is a popular technology for this problem. However, the initiation of the alignment strongly affects the performance of ASM. If the initiation of alignment is bad, the iteration of ASM optimization will be stuck in a local minima, and the alignment will fail. In this paper, we propose a novel approach to improve ASM by building the classifiers of the face components. We design the SVM classifiers for eyes, mouth and nose, and we use Speeded Up Robust Features(SURF) and Local Binary Pattern(LBP) feature to describe the components which are discriminative for the components than Haar-like features. The face components are firstly located by the classifiers and they indicate the initiation of the alignment. Our approach can make the iterations of ASM optimization converge fast and with the less errors. We evaluate our approach on the frontal views of upright faces of IMM dataset. The experimental results have shown that our approach outperforms the original ASM in terms of efficiency and accuracy.
Authors and Affiliations
Yanyun Qu , Tianzhu Fang , Yanyun Cheng, , , Han Liu
Keywords
Related Articles
- EP ID EP103131
- DOI -
- Views 81
- Downloads 0
OWLMap: Fully Automatic Mapping of Ontology into Relational Database Schema
Semantic web is becoming a controversial issue in current research era. There must be an automated approach to transform ontology constructs into relational database so that it can be queried efficiently. The previous re...
Semantic Retrieval Approach for Web Documents
Because of explosive growth of resources in the internet, the information retrieval technology has become particularly important. However the current retrieval methods are essentially based on the full text matching of k...
Developing an Adaptive Language Model for Bahasa Indonesia
A language model is one of the important compo-nents in a speech recognition system. It is commonly developed using a statistical method called n-gram. However, a standard n-gram cannot be used for general domains with s...
Insights on Error-Resilient Image Transmission Schemes on Wireless Network
Usage of image as data (or signal) is quite frequent in majority of the user-centric application. However, transmission of image over non-concrete communication medium like air is still vulnerable due to inherent weaknes...
Performance Comparison between Merge and Quick Sort Algorithms in Data Structure
In computer science field, one of the basic operation is sorting. Many sorting operations use intermediate steps. Sorting is the procedure of ordering list of elements in ascending or descending with the help of key valu...