Camera Motion Control for Scientific Volume Data Animation
The need to visualize the findings has become a vital part for scientists. Various tools are already being used for validating and exploring the data. There are visualization softwares that can be used to get lifelike images. But they are not advisable to use when it comes to scientific researches which are usually complex data. So animating such complicated structures and indefinite spatial relationships is more feasible. On the other hand creating the complex animation is out the scope of scientists as it requires expert animators and the reliable tools. This is because apart from the basic steps of creating animation using keyframe interpolation and choosing the right views, further adjustments in case of complicated parameters like actual camera path, depends upon professional tools that are time consuming for scientists to easily use. A solution to the camera motion design and path generation problem for making volume data animations is presented by Wei-Hsien Hsu, Yubo Zhang and Kwan-Liu Ma .
Camera control is a nontrivial problem in computer graphics . A simple camera model has at least six degrees of freedom, moreover difficulty of setting up a view sequence to efficiently connect several points of interest (POIs).). Number of methods has been advanced to fulfil the needs in video-games, computer games, robotics and cinemas. However, camera motion control for animations has been neglected in the case of scientific visualization. Since scientists are presently working with increasingly large-scale datasets and need to observe features in complex volume data, camera motion path planning for navigation and presentation of volume data is as important as volume classification, feature identification, and rendering .
Complex 3D data which involves different materials or intensities can be effectively represented by volume rendering. It has become a primary scientiﬁc visualization tool. Different colours and opacities are assigned to features by defining transfer functions. After classification is complete and with the support of interactive tools, locating interesting features in the data and picking good views to further examine the volume are generally easy tasks for scientists. However, designing camera movements and creating animations to present volume features is a totally different thing to do. The interpolated views for transitioning from one feature to another in the traditional keyframe-based approach can become problematic because keyframe can lead wandering in the spatial domain. This is especially true when two keyframe have different views i.e. positions and directions. Moreover, in volume visualization, camera paths may penetrate opaque volumetric regions or the view may end up pointing away from the main feature. This results in disorientation for viewers causing loss of information. Such situations can be handled...