Deaktivierte Dienste WebGL in GIScience
This website presents examples of visualization and analysis of different geodata types using the new HTML5 WebGL technology. To see the new possibilities of webGIS applications using WebGL just have a look at the examples below.
All you needed to watch it was an WebGL enabled browser (runs best on Mozilla Firefox).
In Firefox type "about:config" in the addressline and set the property "webgl.force-enabled" to true by doubleclicking.
GIS Analysis with WebGL
This example shows a browser side real-time hillshade calculation of a 1024x1024 input DEM raster. It uses WebGL technology to exploit the parallel computation power of the graphics card (GPU). That way the application computes the value of each pixel in parallel instead of classical serial computation on the main processor (CPU). You can also watch the DEM used for the calculation, SLOPE and ASPECT calculations which are a byproduct of the HILLSHADE algorithm. Try also the difference between 8 and 16 bit DEM precision.
WebGL Point Cloud Viewer
Visualize your point cloud data directly with this viewer from your local disk. No upload is needed, as this application uses the new HTML5 File API.
OSM Examples in your Web Browser
This website section presents examples of OSM visualizations using the new HTML5 WebGL technology.
GPS data uploaded to the OpenStreetMap is collected by the project contributors primarily to support mapping. As the inclusion of roads and paths in the map is a major goal of the project, the collected GPS data often follows such man-made structures. We have extracted all points from the planet.gpx file and visualise the subset that has elevation information in Heidelberg as a 3D point cloud. There are ~1,000,000 individual GPS points in Heidelberg (planet file from 04/2013). The bounding box that was applied has a size of ~200km².
Point Clouds in your Web Browser
This website section presents examples of point clouds visualizations using the new HTML5 WebGL technology.
Dechen-Cave (Iserlohn): Pulpit-Grotto
Visit one of the oldest public speleothem caves of Germany, the Dechen-Cave (German: Dechenhöhle, 51° 21' 55.1" N 7° 38' 28.7" E)!
The point cloud covers the 'Pulpit-Grotto' (German: Kanzelgrotte) and was captured with a Riegl VZ-400 terrestrial laser scanner from 7 scan positions on 30th of May 2013. The group of small stalgmites that correspond to the 'Pulpit'-stalagtites was reconstructed in 3D via structure from motion [1,2,3] based on 20 images of an uncalibrated compact camera (Olympus VG110, D700). The resulting point cloud was fused with the laser scanning data in order to fill shaded areas.
[1] Changchang Wu, "VisualSFM: A Visual Structure from Motion System", http://ccwu.me/vsfm/, 2011
[2] Changchang Wu, Sameer Agarwal, Brian Curless, and Steven M. Seitz, "Multicore Bundle Adjustment", CVPR 2011
[3] Rothermel, M., Wenzel, K., Fritsch, D., Haala, N. (2012). SURE: Photogrammetric Surface Reconstruction from Imagery. Proceedings LC3D Workshop, Berlin, December 2012
Reconstruction of a traditional Nubian house, Assuan
The 3D point cloud of a reconstruction of a traditional Nubian house (24°04'42.65''N, 32°53'23.99''E) at the Nubian Museum in Assuan, Egypt, was generated by structure from motion (SFM) [1,2] using 51 uncalibrated camera images (Olympus VG110, D700) taken on 20-01-2013.
The resulting colored 3D point cloud can be visualized in our WebGL point cloud viewer.
[1] Changchang Wu, "VisualSFM: A Visual Structure from Motion System", http://ccwu.me/vsfm/, 2011
[2] Changchang Wu, Sameer Agarwal, Brian Curless, and Steven M. Seitz, "Multicore Bundle Adjustment", CVPR 2011
Structure from Motion (SFM)
The Citadel of Qaitbay in Alexandria (31°12'50.5''N 29°53'8''E) has been subject of testing 3D reconstruction with ordinary uncalibrated camera images using structure from motion (SFM).
VisualSFM [1,2] has been applied for coarse reconstruction, SURE [3] was used for a dense reconstruction of the southern and western facade, subsampling of the dense cloud and registration of the two clouds was done with CloudCompare [4].
The resulting 3D point cloud can be visualized in our WebGL point cloud viewer.
[1] Changchang Wu, "VisualSFM: A Visual Structure from Motion System", http://ccwu.me/vsfm/, 2011
[2] Changchang Wu, Sameer Agarwal, Brian Curless, and Steven M. Seitz, "Multicore Bundle Adjustment", CVPR 2011
[3] Rothermel, M., Wenzel, K., Fritsch, D., Haala, N. (2012). SURE: Photogrammetric Surface Reconstruction from Imagery. Proceedings LC3D Workshop, Berlin, December 2012
[4] http://www.danielgm.net/cc/
3D Scan of the LRG Team
This example shows our LiDAR Research Group.
Point Clouds in the Web
This example shows one of our seminar rooms, at the Institute of Geography, University of Heidelberg.
The data was acquired with a Riegl VZ-400 sensor system.
3D Maize Plants - May and October
Two multitemporal LiDAR campaigns of a maize crop stand have been performed. Here you can view the campaign in May and October. The third image shows both campaigns merged in one view using our WebGL viewer.
The data was acquired with a Riegl VZ-400 sensor system.
