A serious problem in learning probabilistic models is the presence of hidden variables. These variables are not observed, yet interact with several of the observed variables. Dete...
This paper presents an algorithm for learning the time-varying shape of a non-rigid 3D object from uncalibrated 2D tracking data. We model shape motion as a rigid component (rotat...
Lorenzo Torresani, Aaron Hertzmann, Christoph Breg...
Since most real-world applications of classification learning involve continuous-valued attributes, properly addressing the discretization process is an important problem. This pa...
We exploit the biconvex nature of the Euclidean non-negative matrix factorization (NMF) optimization problem to derive optimization schemes based on sequential quadratic and secon...
Abstract-- Extending traditional models for discriminative labeling of structured data to include higher-order structure in the labels results in an undesirable exponential increas...