IRW: An Incremental Representation for Image-Based Walkthroughs

IRW is a new representation for image-based interactive walkthroughs. The target applications for IRW are those that reconstruct a scene from novel viewpoints using samples from a spatial image dataset collected from a plane at eye-level. These datasets consist of pose augmented 2D images and often have a very large number of samples. The IRW representation exploits spatial coherence and rearranges the input samples as epipolar images. The base unit corresponds to a column of the original image that can be individually addressed and accessed. The overall representation supports incremental updates, efficient encoding, scalable performance, and selective inclusion used by different reconstruction algorithms. Evaluations have demonstrated the performance of the IRW representation on a synthetic as well as a real-world environment.

Columns are re-organized along linear epipolar planes to facilitate efficient encoding.

This example shows the re-organized columns which we use as the base element in the IRW encoding scheme.

Publications

David Gotz, Ketan Mayer-Patel, and Dinesh Manocha. IRW: An Incremental Representation for Image-Based Walkthroughs. ACM Multimedia, Juan-les-Pins, France (2002).
[PDF, 1,721k]