Linear subspace methods that provide sufficient reconstruction of the data, such as PCA, offer an efficient way of dealing with missing pixels, outliers, and occlusions that often ...
We present two discriminative methods for name transliteration. The methods correspond to local and global modeling approaches in modeling structured output spaces. Both methods d...
Discriminative methods have shown significant improvements over traditional generative methods in many machine learning applications, but there has been difficulty in extending th...
We demonstrate that log-linear grammars with latent variables can be practically trained using discriminative methods. Central to efficient discriminative training is a hierarchi...
Many approaches to object recognition are founded on probability theory, and can be broadly characterized as either generative or discriminative according to whether or not the dis...
Appearance features are good at discriminating activities in a fixed view, but behave poorly when aspect is changed. We describe a method to build features that are highly stable u...
We present an efficient method for learning part-based object class models from unsegmented images represented as sets of salient features. A model includes parts' appearance...
Flat appearance-based systems, which combine clever image representations with standard classifiers, might be the most effective way to recognize objects using current technologie...
Many approaches to object recognition are founded on probability theory, and can be broadly characterized as either generative or discriminative according to whether or not the di...