Discriminative training for language recognition has been a key tool for improving system performance. In addition, recognition directly from shifted-delta cepstral features has p...
We investigate the task of 2D articulated human pose estimation in unconstrained still images. This is extremely challenging because of variation in pose, anatomy, clothing, and i...
VoDKA is a Video-on-Demand server developed using Erlang/OTP. In this paper, the evolution of the core architecture of the system, designed for improving resource modelling, is de...
Abstract. This paper describes a method of adapting a domain-independent HPSG parser to a biomedical domain. Without modifying the grammar and the probabilistic model of the origin...
Phrase level translation models are effective in improving translation quality by addressing the problem of local re-ordering across language boundaries. Methods that attempt to f...