Computer Graphics is increasingly using techniques from Machine Learning. The trend is motivated by several factors, but the difficulties and expense of modelling is a major driving force. Here 'modelling' is used very broadly to include models of reflection (learn the BRDF of a real material), animation (learn the motion of real objects), as well as three-dimensional models (learn to model complex things). Building around a few examples, we will explore the whys and hows of Machine Learning within Computer Graphics. The course will outline the basics of data-driven modelling, introduce the foundations of probability and statistics, describe some useful distributions, and differentiate between ML and MAP problems. The ideas are illustrated using examples drawn from previous SIGGRAPHs; we'll help non-artists to draw, animate traffic flow from sensor data, and model moving trees from video. 2014 Copyright held by the Owner/Author.
|Title of host publication||ACM SIGGRAPH 2014 Courses, SIGGRAPH 2014|
|Publisher||Association for Computing Machinery|
|Publication status||Published - 2014|
|Event||ACM Special Interest Group on Computer Graphics and Interactive Techniques Conference, SIGGRAPH 2014 - Vamcouver , Canada|
Duration: 10 Aug 2014 → 14 Aug 2014
|Conference||ACM Special Interest Group on Computer Graphics and Interactive Techniques Conference, SIGGRAPH 2014|
|Period||10/08/14 → 14/08/14|