Delivering digitally a realistic appearance of materials is one of the most difficult tasks of computer vision. Accurate representation of surface texture can be obtained by means ...
We propose a domain specific model for statistical machine translation. It is wellknown that domain specific language models perform well in automatic speech recognition. We show ...
The relentless push for high performance in custom digital circuits has led to renewed emphasis on circuit optimization or tuning. The parameters of the optimization are typically...
We present a general framework to incorporate prior knowledge such as heuristics or linguistic features in statistical generative word alignment models. Prior knowledge plays a ro...
Adaptive techniques like voltage and frequency scaling, process variations and the randomness of input data contribute signi cantly to the statistical aspect of contemporary hardwa...