Richard R. Pawlicki

Omnistereo for panoramic virtual environment display systems

This paper discusses the use of omnidirectional stereo for panoramic virtual environments. It presents two methods for real-time rendering of omnistereo images. Conventional perspective stereo is correct everywhere in the visual field, but only in one view direction. Omnistereo is correct in every view direction, but only in the center of the visual field, degrading in the periphery. Omnistereo images make it possible to use wide field of view virtual environment display systems-like the CAVE/spl trade/-without head tracking, and still show correct stereoscopic depth over the full 360/spl deg/ viewing circle. This allows the use of these systems as true multi-user displays, where viewers can look around and browse a panoramic scene independently. Because there is no need to rerender the image according to view direction, we can also use this technique to present static omnistereo images, generated by offline rendering or real image capture, in panoramic displays. We have implemented omnistereo in a four-sided CAVE/spl trade/ and in a 240/spl deg/ i-Con/spl trade/ curved screen projection system. Informal user evaluation confirms that omnistereo images present a seamless image with correct stereoscopic depth in every view direction without head tracking.

Size-Constancy in the CAVE

The use of virtual environments (VE) for many research and commercial purposes relies on its ability to generate environments that faithfully reproduce the physical world. However, due to its limitations the VE can have a number of flaws that adversely affect its use and believability. One of the more important aspects of this problem is whether the size of an object in the VE is perceived as it would be in the physical world. One of the fundamental phenomena for correct size is size-constancy, that is, an object is perceived to be the same size regardless of its distance from the observer. This is in spite of the fact that the retinal size of the object shrinks with increasing distance from the observer. We examined size-constancy in the CAVE and found that size-constancy is a strong and dominant perception in our subject population when the test object is accompanied by surrounding environmental objects. Furthermore, size-constancy changes to a visual angle performance (i.e., object size changed with distance from the subject) when these surrounding objects are removed from the scene. As previously described for the physical world, our results suggest that it is necessary to provide surrounding objects to aid in the determination of an objects depth and to elicit size-constancy in VE. These results are discussed regarding their implications for viewing objects in projection-based VE and the environments that play a role in the perception of object size in the CAVE.

Example-based conceptual styling framework for automotive shapes

Conceptual design in the automotive industry is a time-consuming process. Iterations between concept sketches, created with traditional two dimensional methods, and 3D digital representations of a prototype are currently one of the big bottlenecks. In this paper we present a framework for an integrated 2D-3D design environment. The core of the framework is a model representing the characteristic lines of automotive shapes built from a set of example shapes. From every example shape we extract the same set of characteristic lines and represent them with a feature vector of deformation gradients. Given a set of constraints, our method can generate a new feature vector with an optimization procedure. We provide examples for meaningful manipulations. We demonstrate that these manipulations are intuitive and create plausible shapes.

A sketching interface for feature curve recovery of free-form surfaces

In this paper, we present a semi-automatic approach to efficiently and robustly recover the characteristic feature curves of a given free-form surface. The technique supports a sketch-based interface where the user just has to roughly sketch the location of a feature by drawing a stroke directly on the input mesh. The system then snaps this initial curve to the correct position based on a graph-cut optimization scheme that takes various surface properties into account. Additional position constraints can be placed and modified manually which allows for an interactive feature curve editing functionality. We demonstrate the usefulness of our technique by applying it to a practical problem scenario in reverse engineering. Here, we consider the problem of generating a statistical (PCA) shape model for car bodies. The crucial step is to establish proper feature correspondences between a large number of input models. Due to the significant shape variation, fully automatic techniques are doomed to failure. With our simple and effective feature curve recovery tool, we can quickly sketch a set of characteristic features on each input model which establishes the correspondence to a pre-defined template mesh and thus allows us to generate the shape model. Finally, we can use the feature curves and the shape model to implement an intuitive modeling metaphor to explore the shape space spanned by the input models.

Pushing the Limits of Vehicle Design: Utilizing a Parametric Shape Grammar to Explore Cross-Over Vehicle Concepts

As consumers express a desire for vehicles that are useful for multiple purposes, automobile manufacturers are responding with vehicles that blur the distinctions between vehicle classes. By understanding the traditional classes, new cross-over vehicles can be developed. Parametric shape grammars offer the flexibility to create innovative new designs both within and across classes, while staying within appropriate form and functional constraints attributed to automobiles. Through the use of universal rules and the understanding of allowable parametric ranges, many novel cross-over vehicle concepts can easily be created.Copyright