Abstract. Obtaining ground truth for hyperspectral data is an expensive task. In addition, a number of factors cause the spectral signatures of the same class to vary with location...
Complex simulations can generate very large amounts of data stored disjointly across many local disks. Learning from this data can be problematic due to the difficulty of obtainin...
John Nicholas Korecki, Kevin W. Bowyer, Larry O. H...
This paper proposes a data driven image segmentation algorithm, based on decomposing the target output (ground truth). Classical pixel labeling methods utilize machine learning al...
We study in this paper the problem of bridging the semantic gap between low-level image features and high-level semantic concepts, which is the key hindrance in content-based imag...
We propose a max-margin formulation for the multi-label classification problem where the goal is to tag a data point with a set of pre-specified labels. Given a set of L labels, a...
Bharath Hariharan, Lihi Zelnik-Manor, S. V. N. Vis...