Abstract

Capturing and rendering real world objects and scenes with high visual quality has been one of the main topics in Computer Vision and Graphics in the last decades. Often, we do not only wish to display the captured content but also modify it, interact with it and experience it in an immersive way. Classic Computer Graphics modelling and rendering provides the freedom of modification and animation, but often lacks visual realism, especially if real-time constraints should be met. Also, modelling often involves enormous manual work and rendering and animation require costly physical simulations. Recent approaches directly capture real world objects and scenes and infer object characteristics from the captured data. Furthermore, more and more image-based modeling techniques have been developed in order to meet both the requirements of realistic appearance and animation, especially for complex objects. This is where Computer Vision and Graphics meet. In this talk, I will give an overview on ongoing works covering the whole processing chain from capturing, image and video analysis and understanding to image-based modelling, editing and rendering.

© 2016 Optical Society of America