Learn More
We present a novel and flexible approach to the problem of feature selection, called grafting. Rather than considering feature selection as separate from learning, grafting treats the selection of suitable features as an integral part of learning a predictor in a regularized learning framework. To make this regularized learning process sufficiently fast for(More)
In the standard feature selection problem, we are given a fixed set of candidate features for use in a learning problem, and must select a subset that will be used to train a model that is " as good as possible " according to some criterion. In this paper, we present an interesting and useful variant, the online feature selection problem, in which, instead(More)
Batch implementations of support vector regression (SVR) are inefficient when used in an on-line setting because they must be retrained from scratch every time the training set is modified. Following an incremental support vector classification algorithm introduced by Cauwenberghs and Poggio (2001), we have developed an accurate on-line support vector(More)
We describe the implementation and performance of a genetic algorithm (GA) which generates image feature extraction algorithms for remote sensing applications. We describe our basis set of primitive image operators and present our chromosomal representation of a complete algorithm. Our initial application has been geospatial feature extraction using(More)
Figure 1: Selective Refinement. The headrest is an object of African cultural heritage. In each frame the selected region is refined progressively. Abstract We present a framework for real-time view-dependent refinement, and adapt it to the task of browsing large model repositories on the Internet. We introduce a novel hierarchical representation of atomic(More)
We consider the problem of pixel-by-pixel classification of a multi-spectral image using supervised learning. Conventional supervised classification techniques such as maximiun likelihood classifcation and less conventional ones such as neural networks, typically base such classifications solely on the spectral components of each pixel. It is easy to see(More)
Multi-instrument data sets present an interesting challenge to feature extraction algorithm developers. Beyond the immediate problems of spatial co-registration, the remote sensing scientist must explore a complex algorithm space in which both spatial and spectral signatures may be required to identify a feature of interest. We describe a genetic(More)
Initial estimates of forest damage included 17,000 acres (6,900 ha) of 70-100% tree mortality. Restoration efforts following the fire were complicated by the large scale of the fire, and by the presence of extensive natural and man-made hazards. These conditions forced a reliance on remote sensing techniques for mapping and classifying the burn region.(More)