von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification

Abstract

A number of pattern recognition tasks, e.g., face verification, can be boiled down to classification or clustering of unit length directional feature vectors whose distance can be simply computed by their angle. In this paper, we propose the von Mises-Fisher (vMF) mixture model as the theoretical foundation for an effective deep-learning of such directional… (More)

Cite this paper

@article{Hasnat2017vonMM, title={von Mises-Fisher Mixture Model-based Deep learning: Application to Face Verification}, author={Md. Abul Hasnat and Julien Bohn{\'e} and Jonathan Milgram and St{\'e}phane Gentric and Liming Chen}, journal={CoRR}, year={2017}, volume={abs/1706.04264} }