A highly parallel design of image surface layout recovering on GPGPU

Guan Ru Li*, Bo-Cheng Lai

*此作品的通信作者

    研究成果: Conference contribution同行評審

    摘要

    Surface layout recovering helps computers understand the intricate information in an image by assigning local segments to different geometric classes. It greatly reduces the complexity of the following-up image processing and is widely used in various computer vision applications. However, the algorithm walks through every image pixel and imposes intensive computation requirement. Through comprehensive analysis on the execution behavior, this paper identifies significant parallelism inherent in the algorithm. With careful concerns on both multi-threaded software and parallel hardware, the optimized parallel design on a modern GPGPU has reached an average of 10.7X performance enhancement.

    原文English
    主出版物標題2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers
    DOIs
    出版狀態Published - 2012
    事件2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Hsinchu, 台灣
    持續時間: 23 4月 201225 4月 2012

    出版系列

    名字2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers

    Conference

    Conference2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012
    國家/地區台灣
    城市Hsinchu
    期間23/04/1225/04/12

    指紋

    深入研究「A highly parallel design of image surface layout recovering on GPGPU」主題。共同形成了獨特的指紋。

    引用此