4. Future Work

A number of changes can be made to the visibility-extending algorithms to improve their effectiveness. For example, the disappearance-reappearance (growth-shrinkage) problem can be factored into the algorithms so that the object never grows or shrinks inappropriately. In this way, we could produce a size scaling algorithm that is not subject to the interference between the size constancy and linear perspective cues. Furthermore, the FOV distortion algorithm could be geared to only distort the FOV when drawing particular objects. Although the object's position would still be distorted, the remainder of the scene would not be warped. Also, the FOV algorithm could be reworked to take better advantage of the linear perspective cues. The algorithms, as presented in this thesis, are adequate for the OOD application but could be improved further. Thus, they should be viewed as a foundation for future research into improving threshold depth perception.

In addition to improving the models and algorithms in this thesis, the effects of luminance and contrast in a typical HMD on anti-aliasing algorithms could be investigated. That is, the limited brightness and color ranges available in an HMD limits the additional resolution that can be displayed with an anti-aliasing or blurring algorithm. Determining empirical values for contrast and the resulting visual acuity predicted by the Contrast Sensitivity Function would yield a measure of the usability of anti-aliasing algorithms for improving acuity in HMDs.

The OOD simulation provides only one application of the algorithms and methodology presented here. Any VE system that requires an improvement in threshold depth perception or needs to combat the problems of low display resolution should find the models developed here to be very useful. Furthermore, applications outside the domain of VE may also find the information presented to be relevant; for example, night-vision goggles have also been plagued by low pixel resolution and decreased visibility.

On to next section: Conclusion