Abstract | ||
---|---|---|
This paper presents applications of special types of deep neural networks (DNNs) for audio-visual biometrics. A common example is the DBN-DNN that uses the generative weights of deep belief networks (DBNs) to initialize the feature detecting layers of deterministic feed forward DNNs. In this paper, we propose the DBM-DNN that uses the generative weights of deep Boltzmann machines (DBMs) for initialization of DNNs. Then, a softmax layer is added on top and the DNNs are trained discriminatively. Our experimental results show that lower error rates can be achieved using the DBM-DNN compared to the support vector machine (SVM), linear regression-based classifier (LRC) and the DBN-DNN. Experiments were carried out on two publicly available audio-visual datasets: the VidTIMIT and MOBIO. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/BTAS.2015.7358754 | 2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems (BTAS) |
Keywords | Field | DocType |
audio-visual person recognition,deep neural networks,audio-visual biometrics,DBN-DNN,deep belief networks,deterministic feed forward DNN,generative weights,deep Boltzmann machines,DBM,softmax layer,error rates,support vector machine,SVM,linear regression-based classifier,LRC,audio-visual datasets,VidTIMIT,MOBIO | Pattern recognition,Softmax function,Computer science,Deep belief network,Support vector machine,Speech recognition,Artificial intelligence,Deep learning,Biometrics,Initialization,Classifier (linguistics),Artificial neural network | Conference |
ISSN | Citations | PageRank |
2474-9680 | 3 | 0.41 |
References | Authors | |
13 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mohammad Rafiqul Alam | 1 | 8 | 2.54 |
M. Bennamoun | 2 | 3197 | 167.23 |
Roberto Togneri | 3 | 814 | 48.33 |
Ferdous Ahmed Sohel | 4 | 623 | 31.78 |