Date of Award

5-2018

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Graphics Technology

Committee Chair

David M. Whittinghill

Committee Member 1

Yingjie Chen

Committee Member 2

Qifan Song

Abstract

The gaming industry always faces the continual challenges of striking balance between real-time computing performance and graphics presentation quality. This is particularly apparent for a game with vast wild-lands scenes which demands for a large number of tree models. From the visual perception perspective, this paper investigated whether people can distinguish trees with uniqueness of various features (relative orientation, density of branches, leaf color) using the same tree model in a virtual environment. During the controlled experiment, quantitative research methods were used, which measured the statistics of tree variations. The results suggested that our methods successfully contributed a strategy in saving graphic resources without user consciousness and maintaining high graphics performance.

COinS