CS Seminar - Large Object Extraction for Binary Images on the GPU
Jodrey School of Computer Science
Friday, January 29, 2016
Carnegie Hall 113
Large Object Extraction for Binary Images on the GPU
Object ﬁltering by size is a basic task in computer vision. A common way to extract large objects in a binary image is to run the connected-component labeling (CCL) algorithm and to compute the area of each component. Selecting the components with large areas is then straightforward. Several CCL algorithms for the GPU have already been implemented but few of them compute the component area. This extra step can be critical for real-time applications such as real-time video segmentation. After introducing the concept of General Purpose GPU programming, this presentation will present a new approach for the extraction of visually large objects in a binary image that works in real-time. It is implemented using CUDA (Compute Uniﬁed Device Architecture), a parallel computing architecture developed by NVIDIA.
About the Presenter
Grégory Huchet graduated from the École Nationale Supérieure des Télécommunications de Bretagne, Brest, France, in 2004. He received a Ph.D in electrical engineering from Laval University, Quebec, Canada, in 2009 and his thesis was on the subject of Distributed Video Coding. From 2009 to 2011, he was a postdoctoral fellow at the Communications Research Centre, Ottawa, Canada. During that time he contributed on the development of a real time 2D-to-3D video converter. Lastly from 2011 to 2015 he was a staff research engineer at the Digital Media Solutions Laboratory of Samsung Research America in Irvine, California. His work was mainly focused on the improvement of image quality and the complexity reduction of the frame rate-up converter found in Samsung Smart TVs.
Everyone is welcome to attend