|
Title:
|
INTRA-CLASS CONSTRAINED SPARSE
REPRESENTATION WITH GRADIENT FEATURE
FOR OCCLUDED FACE RECOGNITION |
|
Author(s):
|
Qinqin Ma and Xiang Ma |
|
ISBN:
|
978-989-8533-91-3 |
|
Editors:
|
Katherine Blashki and Yingcai Xiao |
|
Year:
|
2019 |
|
Edition:
|
Single |
|
Keywords:
|
Occluded Face Recognition, Intra-Class Constraint |
|
Type:
|
Full Paper |
|
First Page:
|
247 |
|
Last Page:
|
253 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
A novel occlusion face recognition method based on gradient feature and intra-class constrained sparse representation
classification is proposed in this paper. We propose a second-order gradient feature to extract the texture and contour of
images, which is more robust to occlusion than the original pixel of the face image. Intra-class constraint sparse
representation is designed to make the test image close to the specific class by minimizing intra-class error. Furthermore,
intra-class error is used as classification criteria. Experiments on the Extended Yale B and AR face databases prove the
superiority and robustness of the proposed method for occluded face recognition, especially for the real world occlusion. |
|
|
|
|
|
|
