Sciweavers

2901 search results - page 74 / 581
» Predicting non-stationary processes
Sort
View
ICIP
2009
IEEE
15 years 1 months ago
Virtual resolution enhancement of scale invariant textured images using stochastic processes
We present a new method of magnification for textured images featuring scale invariance properties. The procedure preserves the visual aspect as well as the statistical properties...
Emilie Koenig, Pierre Chainais
ICASSP
2011
IEEE
14 years 7 months ago
Motion vector recovery with Gaussian Process Regression
In this paper, we propose a Gaussian Process Regression (GPR) framework for concealment of corrupted motion vectors in predictive video coding of packet video systems. The problem...
Hadi Asheri, Abdolkhalegh Bayati, Hamid R. Rabiee,...
QSIC
2005
IEEE
15 years 9 months ago
Ontology Based Requirements Analysis: Lightweight Semantic Processing Approach
We propose a software requirements analysis method based on domain ontology technique, where we can establish a mapping between a software requirements specification and the doma...
Haruhiko Kaiya, Motoshi Saeki
IDEAL
2005
Springer
15 years 9 months ago
Neural Networks: A Replacement for Gaussian Processes?
Abstract. Gaussian processes have been favourably compared to backpropagation neural networks as a tool for regression. We show that a recurrent neural network can implement exact ...
Matthew Lilley, Marcus R. Frean
EMSOFT
2004
Springer
15 years 9 months ago
Multiple process execution in cache related preemption delay analysis
Cache prediction for preemptive scheduling is an open issue despite its practical importance. First analysis approaches use simplified models for cache behavior or they assume si...
Jan Staschulat, Rolf Ernst