High dynamic environment sequence sampling
Ying Yang, Changguo Yu, Liang Wan, Wei Feng
Proceedings of the 13th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry, 2014, pp. 123--131.
Abstract: In this paper, we propose a novel method for high dynamic environment sequence sampling. Without future frame information, our approach achieves temporal coherence by mapping sampling information from previous frame to current frame based on temporal superpixels. However, temporal superpixels have unbalanced importance for rendering, although light regions in environment sequences can be tracked across frames. To solve this issue, we develop an adaptive merge-and-split scheme to adjust sample segments to achieve a more importance-balanced sample distribution. Compared with several state-of-the-art methods, our approach gets consistently improved rendering quality across environment sequences. Experiment results also show decent temporal coherence of our approach in sequence rendering.
@inproceedings{10.1145-2670473.2670478,
author = {Ying Yang and Changguo Yu and Liang Wan and Wei Feng},
title = {High dynamic environment sequence sampling},
booktitle = {Proceedings of the 13th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and its Applications in Industry},
pages = {123--131},
year = {2014},
}
Return to the search page.
graphbib: Powered by "bibsql" and "SQLite3."