Search

PhotoCloud is a real-time client-server system for interactive visualization and exploration of large datasets comprising thousands of calibrated 2D photographs of a scene and a complex 3D description of the scene. The system isn't tailored to any specific data acquisition process; it aims at generality and flexibility. PhotoCloud achieves scalability through a multiresolution dynamic hierarchical representation of the data, which is remotely stored and accessed by the client through an efficient cache system. The system includes a compact image browser and a multiresolution model renderer. PhotoCloud employs iconic visualization of the images in the 3D space and projects images onto the 3D scene on the fly. Users can navigate the 2D and 3D spaces with smooth, integrated, seamless transitions between them. A study with differently skilled users confirms PhotoCloud's effectiveness and communication power. The Web extras at http://www.youtube.com/playlist?list=PLHJB2bhmgB7cmYD0ST9CEDMRv1JlX4xPH are videos demonstrating PhotoCloud, a real-time client-server system for interactive exploration of large datasets comprising 2D photos and 3D models. [ABSTRACT FROM PUBLISHER]

A new method for interactive rendering of complex lighting effects combines two algorithms. The first performs accurate ray tracing in heterogeneous refractive media to compute high-frequency phenomena. The second applies lattice-Boltzmann lighting to account for low-frequency multiple-scattering effects. The two algorithms execute in parallel on modern graphics hardware. This article includes a video animation of the authors' real-time algorithm rendering a variety of scenes. [ABSTRACT FROM PUBLISHER]

Streamit lets users explore visualizations of text streams without prior knowledge of the data. It incorporates incoming documents from a continuous source into an existing visualization context with automatic grouping and separation based on document similarities. Streamit generates document clusters to promote better understanding. To obtain different clusterings, users can adjust the keyword importance on the fly. Topic modeling represents the documents with higher-level semantic meanings. System performance has been optimized to achieve instantaneous animated visualization even for very large data collections. A powerful user interface allows in-depth data analysis. The video shows an example of applying our system on 1,000 US National Science Foundation Information and Intelligent Systems award abstracts funded between March 2000 and August 2003. The visual layout consists of a main window (left view), an animation control panel (bottom), control tools (top right), a keyword table (middle right), and document tables (bottom right). Documents are represented by pies whose size conveys the project's funding. The example shows how clusters of documents are generated and dynamically evolve (move, split, or merge) as new documents are inserted. The simulation places new documents relatively close to similar ones, creating clusters that each have an assigned color. Clusters maintain their colors, which facilitates the visual tracking of their behavior. However, when the system generates new clusters (for example, a cluster splits into two or more clusters), it assigns them unique colors to ease the visual tracking of them as they evolve. For example, in the video, the section from 00:21 to 00:25 shows how the red cluster splits into two clusters: a cluster that keeps the same red color and a new light-blue cluster. Finally, the spiral view (00:32–00:35) lets users examine the clusters' temporal trends. [ABSTRACT FROM PUBLISHER]

Real-time rendering of large-scale vector maps over terrain surfaces requires displaying substantial numbers of polylines and polygons. Because a long latency in display and manipulation is fatal to maintaining presence in a virtual environment, a new method efficiently simplifies and renders such maps. This method consists of three steps. First, it simplifies the vector map while maintaining the map's topological consistency and preventing local conflicts such as intersections or self-intersections. Second, it generates view-dependent level-of-detail (LOD) models. Finally, it overlays the maps onto multiresolution terrain models through the stencil shadow volume algorithm and other techniques. Experiments demonstrated the method's efficiencies in real-time rendering of large-scale vector maps over various LOD terrain surfaces. [ABSTRACT FROM PUBLISHER]