Join GitHub today
GitHub is home to over 31 million developers working together to host and review code, manage projects, and build software together.Sign up
a tool for visualising 3D scenes reconstructed from photographs
Fetching latest commit…
Cannot retrieve the latest commit at this time.
|Type||Name||Latest commit message||Commit time|
|Failed to load latest commit information.|
ABOUT PixelStruct is an opensource tool for visualizing 3D scenes reconstructed from photographs. It uses the opensource structure-from-motion system Bundler, which is based on the same research as Microsoft Live Labs Photosynth. HOW TO COMPILE PixelStruct requires the following libraries: - Qt4 (with OpenGL support) - CGAL You will also need to install CMake in order to build PixelStruct. All of the above can be installed in Ubuntu (and likely other Debian-based distributions) by running sudo aptitude install libqt4-dev libcgal-dev cmake Once you have installed these, compile the program by running `make` in the project root directory. This will create a binary called 'pixelstruct' in the current directory. You can either copy this executable into your PATH for system-wide access, or just run it immediately by either calling `./pixelstruct` in a terminal or by opening it with your file browser. USAGE PixelStruct requires Bundler, which can be downloaded from . First, you need to manually run Bundler on a directory of images as described in the README.txt file contained in the Bundler distribution. Once Bundler has completed processing the images, the image set can be loaded into PixelStruct. If you do not want to manually run Bundler, and would rather view a pre-processed dataset, you have two options: - download the small (3.6MB) Kermit example from , untar it, and open the kermit/ directory in PixelStruct - download the very large (1.8GB) Notre Dame example from , unzip it, create a bundle/ directory inside the NotreDame/ directory, copy/move notredame.out to bundle/bundle.out, and open the NotreDame/ directory in PixelStruct To view a processed dataset, launch PixelStruct as described above, and go to "File>Open image directory". Navigate into the image directory you ran Bundler on, and hit "Choose"/"Open". Alternatively you can specify the image directory as a command-line argument when launching PixelStruct. The 3D point cloud will be loaded into the window, and once the viewer has focus (you may have to click on it for this), the camera will focus on the first image. You can now hit your Return key to move to the next image, or use your arrow keys to move to another image in the chosen direction. Pressing Page Up and Page Down will attempt to move the camera closer to and further away from the photographed object respectively. The View menu allows you to choose between several transition modes, as described in section 5.3 of . QPL-GPL INCOMPATIBILITY NOTICE As portions of CGAL are licensed under the QPL which is incompatible with the GPL (see the QPL secction of ), the following notice is required: As a special exception, you have permission to link this program with the CGAL library and distribute executables, as long as you follow the requirements of the GNU GPL in regard to all of the software in the executable aside from CGAL.  http://da.vidr.cc/projects/pixelstruct/  http://phototour.cs.washington.edu/bundler/  http://photosynth.net/  http://pixelstruct.googlecode.com/files/kermit-example-bundler0.3.tar.gz  http://phototour.cs.washington.edu/datasets/NotreDame.zip  http://phototour.cs.washington.edu/Photo_Tourism.pdf  http://doc.trolltech.com/4.3/qpl.html  http://www.fsf.org/licensing/licenses/index_html