Rendering by Manifold Hopping

Perceptual image-based rendering using a small set of multi-perspective images

Project Description

In this paper, we present a novel image-based rendering technique, which we call manifold hopping. Our technique provides users with perceptually continuous navigation by using only a small number of strategically sampled manifold mosaics or multi-perspective panoramas. Manifold hopping has two modes of navigation: moving continuously along any manifold, and discretely between manifolds. An important feature of manifold hopping is that significant data reduction can be achieved without sacrificing output visual fidelity, by carefully adjusting the hopping intervals. A novel view along the manifold is rendered by locally warping a single manifold mosaic using a constant depth assumption, without the need for accurate depth or feature correspondence. The rendering errors caused by manifold hopping can be analyzed in the signed Hough ray space. Experiments with real data demonstrate that we can navigate smoothly in a virtual environment with as little as 88k data compressed from 11 concentric mosaics.

Publications

·         Rendering by Manifold Hopping, International Journal of Computer Vision, November 2002

Project Team

  • Jinxiang Chai (Carnegie Mellon University)
  • Harry Shum (Microsoft Research Asia)
  • Lifeng Wang (Microsoft Research Asia)
  • Xin Tong (Microsoft Research Asia)

Related projects

·         Plenoptic Sampling,  in Computer Graphics (ACM SIGGRAPH 2000)

·        Layered lumigraph with LOD control. Journal of Visualization and Computer Animation 13(4): 249-261 (2002).


Jinxiang Chai
Last Updated: May 10, 2004