Speaker: Prof. Paul L. Rosin
Date and Time: 9:00 - 10:00 am, Friday, August 14, 2015
Venue: Lecture Hall, Second Floor, Office Building, Software Campus
Title: patch queries for large datasets
Non-Photorealistic Rendering of Images Computer Scientists have been inspired by artists to develop many non-photorealistic rendering techniques that can take an input image and produce a wide variety of artistic effects such as oil painting, watercolour, pointillist, etc. Their application can simply be for fun (e.g. to make consumer photographs more interesting), but there are also many other applications such as scientific visualisation, film production effects, technical illustration, etc. My talk will overview several NPR techniques developed by myself and Yu-un Lai , that aim to generate results that are relatively minimal rendering, in terms of using a combination of refined lines and regions and a small number of tones or colours.The aim is to achieve a rendering that is abstracted but retains sufficient elements from the original image so that (for example) it can be used to generate recognisable portraits.
Paul L. Rosin is a Professor at the School of Computer Science & Informatics, Cardiff University. Previous posts include Brunel University, Joint Research Centre, Italy and Curtin University of Technology, Australia. His research interests include shape analysis, curve segmentation, low level image processing, machine vision approaches to remote sensing, medical and biological image analysis,mesh processing, non-photorealistic rendering, the analysis of shape in art and architecture, and application of cellular automata to image processing.