We develop a framework for learning generic, expressive image priors that capture the statistics of natural scenes and can be used for a variety of machine vision tasks. The appro...
Blur from camera shake is mostly due to the 3D rotation of the camera, resulting in a blur kernel that can be significantly non-uniform across the image. However, most current de...
Oliver Whyte, Josef Sivic, Andrew Zisserman, Jean ...
Abstract. Self-efficacy is an individual's belief about her ability to perform well in a given situation. Because selfefficacious students are effective learners, endowing int...
Scott W. McQuiggan, Bradford W. Mott, James C. Les...
—One of the major challenges in multi-view imaging is the definition of a representation that reveals the intrinsic geometry of the visual information. Sparse image representati...
We present a global approach that enables the production of 3D soccer sequences from non-calibrated video cameras. Our system can produce a 3D animated model of the scene from a si...