NIPS
2004
14 years 25 days ago
2004
We present a generative model and stochastic filtering algorithm for simultaneous tracking of 3D position and orientation, non-rigid motion, object texture, and background texture...
NIPS
2004
14 years 25 days ago
2004
While clustering is usually an unsupervised operation, there are circumstances in which we believe (with varying degrees of certainty) that items A and B should be assigned to the...
NIPS
2004
14 years 25 days ago
2004
Visual action recognition is an important problem in computer vision. In this paper, we propose a new method to probabilistically model and recognize actions of articulated object...
NIPS
2004
14 years 25 days ago
2004
We establish a mistake bound for an ensemble method for classification based on maximizing the entropy of voting weights subject to margin constraints. The bound is the same as a ...
NIPS
2004
14 years 25 days ago
2004
We propose a "soft greedy" learning algorithm for building small conjunctions of simple threshold functions, called rays, defined on single real-valued attributes. We al...
NIPS
2004
14 years 25 days ago
2004
This paper concerns approximate nearest neighbor searching algorithms, which have become increasingly important, especially in high dimensional perception areas such as computer v...
NIPS
2004
2004
Methods for Estimating the Computational Power and Generalization Capability of Neural Microcircuits
14 years 25 days ago
What makes a neural microcircuit computationally powerful? Or more precisely, which measurable quantities could explain why one microcircuit C is better suited for a particular fa...
NIPS
2004
14 years 25 days ago
2004
Multiple realizations of continuous-valued time series from a stochastic process often contain systematic variations in rate and amplitude. To leverage the information contained i...
NIPS
2004
14 years 25 days ago
2004
This paper presents an adaptive discriminative generative model that generalizes the conventional Fisher Linear Discriminant algorithm and renders a proper probabilistic interpret...
NIPS
2004
14 years 25 days ago
2004
Bayesian Regularization and Nonnegative Deconvolution (BRAND) is proposed for estimating time delays of acoustic signals in reverberant environments. Sparsity of the nonnegative f...