ScanView 1.21

ScanView is a client / server rendering system for viewing complex 3D models, like those produced using laser triangulation scanners, without downloading the models. This lets you interactively examine models that would otherwise be inaccessible due to their large size or due to usage restrictions. In particular, it allows the general public to view the high-resolution 3D models generated during Stanford's Digital Michelangelo Project without obtaining a license.

The client part of ScanView consists of a freely available viewer program and a set of simplified 3D models. Using the viewer, you select one of the simplified models, then position and orient it on your computer screen. Feedback is provided in the form of images rendered from the simplified model. The quality of these image is limited, but it is sufficient for navigation. You can also change the lighting and surface appearance. When you release the mouse, ScanView queries our rendering server (using an HTTP protocol) for an image corresponding to your current view, but rendered from our high-resolution model. This image is sent to your client, overwriting your image of the simplified model. A typical query-response cycle takes 1-2 seconds, but this latency may vary depending on network traffic, the speed of your connection, and the current load on our server.

ScanView currently runs on any Windows-based PC, with or without hardware graphics acceleration. It is currently configured to visualize two (normally inaccessible) models from the Digital Michelangelo Project's Archive of 3D Models (David and St. Matthew), one publically accessible fragment of the Forma Urbis Romae, a giant marble map of ancient Rome, and two models from the unrestricted Stanford 3D Scanning Repository (the bunny and the happy buddha).

The ScanView system was developed by David Koller, Michael Turitzin, and Marc Levoy of the Stanford Computer Graphics Laboratory. See the bottom of this web page for the copyright notice and usage restrictions. The ScanView system is described in a SIGGRAPH 2004 paper by us and Marco Tarini, Giuseppe Croccia, Paolo Cignoni, and Roberto Scopigno. A shortened version of this paper was the cover article in the June 2005 issue of CACM.