Multi-view face and eye detection using discriminant features
Computer Vision and Image Understanding
A visual approach for driver inattention detection
Pattern Recognition
High-Performance Rotation Invariant Multiview Face Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Face refinement through a gradient descent alignment approach
VisHCI '06 Proceedings of the HCSNet workshop on Use of vision in human-computer interaction - Volume 56
Robust facial feature tracking under varying face pose and facial expression
Pattern Recognition
Facial Action Unit Recognition by Exploiting Their Dynamic and Semantic Relationships
IEEE Transactions on Pattern Analysis and Machine Intelligence
Robust object tracking with a case-base updating strategy
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
FaceSeg: automatic face segmentation for real-time video
IEEE Transactions on Multimedia
EURASIP Journal on Advances in Signal Processing
Rotated haar-like features for face detection with in-plane rotation
VSMM'06 Proceedings of the 12th international conference on Interactive Technologies and Sociotechnical Systems
3D face pose estimation based on multi-template AAM
CCBR'12 Proceedings of the 7th Chinese conference on Biometric Recognition
Hi-index | 0.00 |
In current face detection, mostly often used features are selected from a large set (e.g. Haar wavelets). Generally Haar wavelets only represent the local geometric feature. When applying those features to profile faces and eyes with irregular geometric patterns, the classifier accuracy is low in the later training stages, only near 50%. In this paper, instead of brute-force searching the large feature set, we propose to statistically learn the discriminant features for object detection. Besides applying Fisher discriminant analysis (FDA) in AdaBoost, we further propose the recursive nonparametric discriminant analysis (RNDA) to handle more general cases. Those discriminant analysis features are not constrained with geometric shape and can provide better accuracy. The compact size of feature set allows to select a near optimal subset of features and construct the probabilistic classifiers by greedy searching. The proposed methods are applied to multi-view face and eye detection and achieve good accuracy.