Manuel M. Oliveira
Efficient Edge-Aware Filtering
Abstract
Edge-aware filtering is a fundamental building block to a wide range of image and video processing, and computer graphics applications. These include detail enhancement, denoising, stylization and non-photorealistic effects, recoloring and colorization, tone mapping, and photon-map filtering, just to name a few. While an edge-preserving filter can be implemented as a convolution with a spatially-invariant kernel in high-dimensional space, performing such an operation is computationally expensive, preventing its use in interactive and real-time scenarios. The talk will present two recent techniques we have developed for efficiently performing edge-aware filtering. The first one is based on a domain transform that defines an isometry between curves on 2-D image manifolds (embedded in n-D space) and the real line. High-dimensional geodesic filtering is then performed in linear time as a sequence of 1-D filtering steps using a spatially-invariant kernel. The second technique focuses on Euclidean filtering. It works by sampling and filtering the input signal using a set of 2-D manifolds adapted to the original data. It essentially queries the value of a multivariate function in n-D by interpolating several scattered samples using normalized convolution. Its cost is linear both in the number of pixels and in the dimensionality of the space in which the filter operates. These techniques have many desirable features. In particular, they are significantly faster than previous approaches, supporting high-dimensional filtering of images and videos in real time. In the talk, I will show several examples illustrating their use in image, video processing, and computer graphics applications.
Short Bio
Manuel M. Oliveira is an Associate Professor of Computer Science at the Federal University of Rio Grande do Sul (UFRGS), in Brazil. He received his PhD from the University of North Carolina at Chapel Hill, in 2000. Before joining UFRGS in 2002, he was an Assistant Professor of Computer Science at the State University of New York at Stony Brook (2000 to 2002). In the 2009-2010 academic year, he was a Visiting Associate Professor at the MIT Media Lab. His research interests cover most aspects of computer graphics, but especially the frontiers among graphics, image processing, and vision (both human and machine). In these areas, he has contributed a variety of techniques including relief texture mapping, real-time filtering in high-dimensional spaces, efficient algorithms for Hough transform, new physiologically-based models for color perception and pupil-light reflex, and novel interactive techniques for measuring visual acuity. His work has been marked by a quest for solutions that produce high-quality results in real time.
Manuel was program co-chair of the ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games 2010 (I3D 2010), and general co-chair of ACM I3D 2009. He was also program co-chair of the WSCG 2013 and SIBGRAPI 2006. He received the ACM Recognition of Service Award in 2009 and 2010.