MPI-INF Logo
Campus Event Calendar

Event Entry

What and Who

Exploring computer vision and machine learning for Model Capturing and Synthesis

Chuan Li
University of Bath
Talk

Chuan Li is a post-doc research officer at University of Bath. He obtained his bachelor degree in software engineering at Zhejiang University (2005) and his PhD degree in computer science at Bath (2010).
AG 1, AG 2, AG 3, AG 4, AG 5, RG1, MMCI  
AG Audience
English

Date, Time and Location

Friday, 22 February 2013
11:00
45 Minutes
E1 4
019
Saarbrücken

Abstract

Creating digital models of urban scenes and natural environment is important to many applications such as digital entertainment and civil planning. This talk will explore low-cost solutions for capturing and synthesizing objects that are traditionally expensive to create/edit. 

First, I will introduce my recent work on video-based modelling for dynamic natural phenomena. I will focus on a probabilistic approach that is able to capture a realistic moving tree from a single view-point video. I will also introduce a low cost solution for acquiring a visually plausible water surface model, again from a single view-point video taken in the natural world.

Second, I will use buildings as the subject to discuss my work on example-based model synthesis. I will introduce 1) a technique that automatically captures the appearance style of an example building and retargets the style to new buildings; 2) a method that automatically synthesizes novel models by re-combining parts from different example buildings.

More details can be found at

http://www.cs.bath.ac.uk/~cl249/research.html

Contact

Michael Wand
4008
--email hidden
passcode not visible
logged in users only

Michael Wand, 02/20/2013 12:20 -- Created document.