High-level generative models provide elegant descriptions of videos and are commonly used as the inference framework in many unsupervised motion segmentation schemes. However, app...
Currently 3D animation rendering and video compression are completely independent processes even if rendered frames are streamed on-the-fly within a client-server platform. In suc...
Robert Herzog, Shin-ichi Kinuwaki, Karol Myszkowsk...
Abstract Weproposeacomputationalmodelofcontourintegration for visual saliency. The model uses biologically plausible devices to simulate how the representations of elements aligned...
This paper presents visualization of field-measured, time-varying multidimensional earthquake accelerograph readings. Direct volume rendering is used to depict the space-time rela...
Background: Natively unfolded proteins lack a well defined three dimensional structure but have important biological functions, suggesting a re-assignment of the structure-functio...