You are here: Home Research and PhD Graphics, Vision, Audio

Graphics, Vision, Audio

Processing of digital image and audio data is a central topic in computer science, as it caters to the most important devices of human communication: vision and hearing. The amount of data like images, videos, sound, speech, motion or 3D models is rapidly growing and even the digitization of the whole earth is under way. Digital abstract data are generated in nearly all disciplines ranging from high dimensional simulation data in chemistry and physics to sequence data in biology and network data in the social media context.

On the one hand, new methods are needed to locate important information in this overwhelming amount of data and present it to the user. Example questions in this context are: how can intelligent learning techniques such as deep learning be employed to efficiently obtain relevant information from mass data? How can complex audio-visual scenes be efficiently monitored by using suitable visual and acoustic object detection and classification? How can the content of billions of images or videos be analyzed? How can we interpret the movements and actions of people in images and video? How can we use natural language to help computers to see? How can a computer understand and forecast human behavior? How can computers interpret medical and scientific images in a quantitative and objective manner?

On the other hand, novel technology and techniques are required to digitize the reality or to generate digital 3D- and audio content for virtual worlds, e.g. for movies, 3D games or visualization purposes. Related questions are: How can computation enable better cameras and novel sensing technologies? How can we digitize, transmit and remotely render dynamic 3D scenes for 3d telepresence? How can we realistically render clouds, ocean waves, woven cloth, skin or hair in a movie or game? How can we describe and represent the variance in shape, motion and appearance collections and how exploit this information to generate new content? How can model-based and data-driven methods be combined to better analyze and synthesize human and animal motions? How can symbolic and numeric approaches be combined for simulation and data analysis? How can computer graphics become a near-physical simulation tool for light from nanometer to kilometer scales? How can interactive visualization help us understand large and complex data? What are intuitive interaction metaphors to navigate in a virtual world?

All these and related questions arise in numerous scientific fields like agriculture, archaeology, biology, chemistry, geology, medicine, physics and social sciences. Answering these questions and deriving new methods and techniques to analyze, manipulate and create the data arising from the wide variety of applications outline above requires foundational as well as applied research. In order to cope with the long list of open challenges in the field, a deep understanding of the underlying theory and algorithms as well as good skills in implementing the relevant methods are indispensable.

The Graphics, Vision and Audio (GVA) track within our MSc program is designed to impart the profound and broad scientific knowledge necessary to work and to research as a computer scientist in the fields of Graphics, Vision and Audio. Excellent job perspectives exist in growth sectors such as the aeronautic and automotive industries, the computer game and film industries, the multimedia industry, e-commerce, social networks and in digital libraries.

Research Groups

Strong research groups conduct outstanding fundamental and applied research projects in the areas of


Document Actions