Image Representation Using 2D Gabor Wavelets
IEEE Transactions on Pattern Analysis and Machine Intelligence
Automatic Classification of Single Facial Images
IEEE Transactions on Pattern Analysis and Machine Intelligence
Face recognition by elastic bunch graph matching
ICIP '97 Proceedings of the 1997 International Conference on Image Processing (ICIP '97) 3-Volume Set-Volume 1 - Volume 1
A Robust Approach to Face and Eyes Detection from Images with Cluttered Background
ICPR '98 Proceedings of the 14th International Conference on Pattern Recognition-Volume 1 - Volume 1
| Hi-index | 0.00 |
This paper presents a technique to identify faces using correspondence of feature points between faces. One to many correspondence mapping among feature points is performed to achieve the best fit among feature points of two given face images. A set of nine feature points is selected during image registration, which represents the approximated landmark points of the set of face images of the registered person. During recognition, a 5脳5 neighborhood of the matching image anchored at each corresponding feature point location in the registered image is searched for the best matching point. As a result, a set of feature vectors for the matching image is secured. Feature vectors are calculated using Gabor responses at these points as they are known to be effective in providing local feature descriptors. The best estimation for matching points and the final face similarity is calculated using the nearest-neighbor algorithm using the Euclidean distance. We evaluate the effectiveness of the feature selection method described in this paper on frontal and near-frontal face images in a large database.