Model-based multi-component volume data labeling and rendering

We present an intuitive multi-component volume data labeling and rendering system, driven by a given annotated semantic model (bought from [PlasticboyCC 2012]). Our system is based on a key idea: a user can obtain a reasonable specification for segmentation, labeling and color mapping by inferring the information from the provided model. This furthermore reduces the degree to which the user needs to be a domain expert, or visualization expert, himself. The input volume data may vary in detail from the model, and it is not annotated with semantic information, so making use of the semantic data still remains a challenging task. Our approach comprises three main steps (see Figure 1).