gpu_nvidia

所属分类:图形图像处理
开发工具:C++
文件大小:7693KB
下载次数:164
上传日期:2007-07-24 01:54:36
上 传 者iseealv
说明:  这是国际上首次采用GPU来实现立体视觉匹配的应用,此后GPU的应用才开始变的广泛起来,很值得从事GPU编程和开发者阅读,里面实现的很精简,读懂要花点功夫哦。
(This is used for the first time the international community to achieve GPU matching stereo vision applications, since the application of GPU began to become widespread, and is worth engaging in GPU programming and developers to read and realize it)

文件列表:
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Include\IOS.H (9468, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Include\IOSTREAM.H (2397, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Include\ISTREAM.H (6487, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Include\OSTREAM.H (5163, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Include\STREAMB.H (5378, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Include\USEOLDIO.H (1078, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Include (0, 2007-07-24)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Lib\MSVCIRTD.LIB (102638, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Lib\STRMBASE.LIB (2081732, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\Lib (0, 2007-07-24)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\System32\msvcirtd.dll (81410, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\System32\MSVCRTD.DLL (385100, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\System32 (0, 2007-07-24)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\1.bmp (331830, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\2.bmp (331830, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\3Ddrome - Tutorial - Cg Pixel Shaders in OpenGL.mht (44469, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\glext.h (153765, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psSAD.txt (841, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psStereoOnePass.txt (894, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psSUM1.txt (869, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psSUM2.txt (1076, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psSUM3.txt (1279, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psSUM4.txt (1382, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psSUM5.txt (1101, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psSUM6.txt (1200, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\psSUM7.txt (1302, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\Thumbs.db (8192, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin\VDTM.exe (94257, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bin (0, 2007-07-24)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bmp_io.cpp (17218, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\bmp_io.h (3228, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\CameraImage.cpp (6571, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\CameraImage.h (4037, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\Debug (0, 2007-07-24)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\glext.h (153765, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\glut32.dll (169984, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\ImageSet.cpp (9622, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\ImageSet.h (4800, 2005-07-05)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\ImageSetView.cpp (44270, 2005-07-23)
第一个采用GPU用于图形加速的源程序(要求显卡支持,nvidia)\vdtm_rect\ImageSetView.h (6267, 2005-07-05)
... ...

********************************************************************** Demo Program for Real-time Stereo on Graphics Hardware, (c) April 2003 By Ruigang Yang (ryang@cs.unc.edu) and Hua Yang (yanghua@cs.unc.edu) For the latest version, please check http://www.cs.unc.edu/~ryang/research/ViewSyn/realtime.htm ********************************************************************** Reference: R. Yang and M. Pollefeys, Multi-Resolution Real-Time Stereo on Commodity Graphics Hardware, in Proc. CVPR 2003. This program demonstrates a real-time implementation of the stereo method in the above reference. For the sake of clarity, this demo program only accepts rectified images. The disparity map is computed over and over for the rectified image pair and the computation time is displayed. The size of the kernel (mipmap levels) can be varied interactively. The full implementation discussed in the paper can use input images in arbitrary configurations without any performance penalty. This software is provided by the contributors "as is" and it comes without any expressed or implied warranties (see the copyright notice in the end). Minimum Requirements to RUN: * A graphic card supporting two texture units and register-combiner (NVIDIA's GeForce2 and above meet the requirement) * Windows 95 and above Minimum Requirements to COMPILE * All the requirements to RUN * GLUT library, freely available from http://www.xmission.com/~nate/glut.html * glext.h, glh_genext.h, wglext.h. They are part of NVIDIA's NVSDK library. NVSDK is huge (over 200 MB). Each file can also be found on google by using the file name as the search keyword. However, there are multiple versions. The authors use files created on 10/20/2002. To avoid changes in the source code, these files should be put in the same directory as gl.h. For MS visual studio users, the default location is "Your VS directory"\VC***\Include\GL. Usage: VDTM LeftImageName RightImageName minDisparityValue maxDisparityValue "LeftImageName" and "RightImageName" are the names of the input images, only Windows BMP files are supported. Images larger than 512x512 are not tested. "minDisparityValue" and "maxDisparityValue" are two integers. They define the disparity search range. Sample Usage: VDTM tsukuba\scene1.1.bmp tsukuba\scene1.0.bmp 0 16 Run-time keyboard binding: *** The OpenGL window must be the active window for input ***** 0-6: The maximum mipmap level (MML) used for summing up the matching scores, default(4) 'c': Toggle between the depth map and the left color image. ESC: Quit the program Issues: * This program has been tested primarily on NVIDIA graphics cards on desktop PCs. It is highly recommended to use NVIDIA's driver 40.72 or later. Some laptops with old NVIDIA drivers have compatibility issues. Thus it is not recommended to run this program on laptops. * The available MML depends on the number of texture units the graphics card supports. The maximum MML is twice of the number of texture units available, i.e., two texture units supports MML up to four, etc. Credit: * bmp_io.cpp and bmpio.h were originally written by John Burkardt. * ocvImage.cpp and ocvImage.h were originally from Intel * Vec4f.hpp is written by Kenny Hoff. * The test images are from the Tsukuba set (courtesy of the University of Tsukuba). April 2003 COPYRIGHT: /*****************************************************************************/ /* Copyright 2003 The University of North Carolina at Chapel Hill. */ /* All Rights Reserved. */ /* Permission to use, copy, modify OR distribute this software and its */ /* documentation for educational, research and non-profit purposes, without */ /* fee, and without a written agreement is hereby granted, provided that the */ /* above copyright notice and the following three paragraphs appear in all */ /* copies. */ /* */ /* IN NO EVENT SHALL THE UNIVERSITY OF NORTH CAROLINA AT CHAPEL HILL BE */ /* LIABLE TO ANY PARTY FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR */ /* CONSEQUENTIAL DAMAGES, INCLUDING LOST PROFITS, ARISING OUT OF THE */ /* USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN IF THE UNIVERSITY */ /* OF NORTH CAROLINA HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH */ /* DAMAGES. */ /* */ /* THE UNIVERSITY OF NORTH CAROLINA SPECIFICALLY DISCLAIM ANY */ /* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF */ /* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE */ /* PROVIDED HEREUNDER IS ON AN "AS IS" BASIS, AND THE UNIVERSITY OF */ /* NORTH CAROLINA HAS NO OBLIGATIONS TO PROVIDE MAINTENANCE, SUPPORT, */ /* UPDATES, ENHANCEMENTS, OR MODIFICATIONS. */ /* */ /* If you have any questions regarding this agreement or the licensing of */ /* this technology, please send them to ryang@cs.unc.edu or marc@cs.unc.edu */ /*****************************************************************************/ ################ Development notes #################### * Testing using rectified images tsukuba\scene1.1.bmp tsukuba\scene1.0.bmp 0 16 * Tesing using un-rectified images full VGA: make sure your screen is wider than 1280 test_00f0003FR.bmp test_01f0003FR.bmp testconfig.txt QVGA: test_00f0003FR_half.bmp test_01f0003FR_half.bmp testconfig.txt * Configuration file format -- comments start with # -- GL_Recon #1 #2 #3 #4 #1: number of depth planes #2: near z-plane (assuming z is the depth) #3: far z-plane #4: scale factor (should always be 1) -- followed by camera calibration information

近期下载者

相关文件


收藏者