Accelerated Algorithm for Silhouette Fur Generation Based on GPU

  • Authors:
  • Gang Yang;Xin-Yuan Huang

  • Affiliations:
  • Beijing Forestry University, Beijing, China 100083 and Institute of Software, Chinese Academy of Sciences, Beijing, China 100080;Beijing Forestry University, Beijing, China 100083

  • Venue:
  • UAHCI '09 Proceedings of the 5th International on ConferenceUniversal Access in Human-Computer Interaction. Part II: Intelligent and Ubiquitous Interaction Environments
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

In the method that represents fur with multi-layer textured slices, representing silhouette fur is a time consuming work, which requires silhouette-edge detection and fin slices generation. In the paper, we present an accelerated method for representing silhouette fur by taking advantage of the programmable ability of Graphic Process Units (GPU). In the method, by appending edge info on each vertex, the silhouette-edge detection can be implemented in GPU; and by storing fin slices data in video memory in preprocessing, the time spent on fin slices generation and on data transmission from CPU to GPU can be saved. Experimental results show that our method accelerates silhouette fur representation greatly, and hence improves the performance of rendering furry objects.