Likelihood normalization for face authentication in variable recording conditions
نویسندگان
چکیده
In this paper we evaluate the effectiveness of two likelihood normalization techniques, the Background Model Set (BMS) and the Universal Background Model (UBM), for improving performance and robustness of four face authentication systems utilizing a Gaussian Mixture Model (GMM) classifier. The systems differ in the feature extraction method used: eigenfaces (PCA), 2-D DCT, 2-D Gabor wavelets and DCT-mod2. Experiments on the VidTIMIT database, using test images corrupted either by an illumination change or compression artefacts, suggest that likelihood normalization has little effect when using PCA derived features, while providing significant performance improvements when using the remaining features.
منابع مشابه
Compensating for variable recording conditions in frontal face authentication algorithms
This paper addresses the problem of compensating for variable recording conditions such as changes in illumination, scale diierences, varying face position. It is well known that the performance of any face authentication/ recognition algorithm deteriorates signiicantly in the presence of the aforementioned conditions as well as the expression variations. The use of simple and powerful pre-proc...
متن کاملIllumination Normalization Using Logarithm Transforms for Face Authentication
In this paper we propose an algorithm that can easily be implemented on small form factor devices to perform illumination normalization in face images captured under various lighting conditions for face verification. We show that Logarithm transformations on images suffering from significant illumination variation, produce face images that are improved substantially for performing face authenti...
متن کاملMorphological elastic graph matching applied to frontal face authentication under well-controlled and real conditions
In this paper, morphological elastic graph matching is applied to frontal face authentication on databases ranging from small to large multimedia ones collected under either well-controlled or real-world conditions. It is shown that the morphological elastic graph matching achieves a very low equal error rate on databases collected under well-controlled conditions. However, its performance dete...
متن کاملMorphological Elastic Graph Matching Applied to Frontal Face Authentication under Optimal and Real Conditions
In this paper, morphological elastic graph matching is applied to frontal face authentication on databases collected either under optimal conditions or during real-world tests (i.e., access-control to buildings or tele-services via INTERNET in a typical office environment etc.). It is demonstrated that the morphological elastic graph matching achieves a very low equal error rate on databases co...
متن کاملQuality-Based Score Normalization for Audiovisual Person Authentication
This paper addresses the problem of biometric audiovisual person authentication in realistic acquisition conditions. Differences in environmental factors or acquisition devices between enrollment and test conditions modify the verification scores distribution and degrade verification performance if not taken into account. A theoretical framework that incorporates quality measures to biometric a...
متن کامل